An Engineer's Ice-Core Thought Experiment

Guest essay by Ronald D Voisin

(For the Ice-Core “Near-Perfect-Recordation” Enthusiasts)

Ice-cores analysis has provided many valuable insights into past climate. We can be more confident for some of these insights than others. One troubling insight surrounds the peaks of historic CO2 atmospheric concentration. In this essay, I would like to describe what I believe would be an “ideal” description of the recordation process regarding the amplitude of the peaks of historic CO2 atmospheric concentration.

Let’s assume that the following atmospheric perturbation is to be recorded in the ice (see Figure 1). At some point in time t0, atmospheric CO2 rises from a background concentration of 300ppm, at 3ppm/year, for 100 years. It peaks at 600ppm representing a 100% spike from the original background concentration and then falls in a similar fashion during the ensuing 100 years. The amplitude and duration of the perturbation are arbitrary but, I think, potentially representative of prior real-world events (and possibly we are just now 60 years into recording the beginning of a somewhat similar event at Mauna Loa currently). So how might this atmospheric perturbation be captured (recorded) accurately in Antarctica ice?

Figures 1&2First let’s describe the initial conditions for the recording media, at t0 and prior, which might be typically found at a drill site in Antarctica (see Figure 2). On the surface there are a couple meters of loose snow blowing around. Under that there is ~100 meters of firn – partially compacted snow that has been left over nominally in layers from previous seasons of snow. The seasonal age of the bottom of the firn is typically 40 years. Under that there is glacial ice going down to great distance formed from the weight pressure of the firn. This glacial ice represents past seasons, nominally in layers, from 40 years to as many as 800,000 years. Atmospheric CO2 is initially at 300ppm.

Figures 3&4In Figure 3 we have a snapshot of the atmospheric CO2 recordation at t0 plus 50 years. During these 50 years the atmospheric CO2 concentration has grown from 300ppm to 450ppm at 3ppm/year. The latter 40 years of the 50 are recorded in the firn while the first 10 years have now progressed into the glacial ice. In Figure 4 another 50 years of recordation have occurred. At t0 plus 100 years, atmospheric CO2 is now peaked at 600ppm. Again the latter 40 years are recorded in the firn while the initial 60 years have now progressed into the glacial ice.

In Figure 5 below we are now on the downside of the atmospheric perturbation to be recorded. Atmospheric CO2 has now fallen from a 600ppm peak back down to 450ppm. As before the most recent 40 years are in the firn; and 110 years of recordation have progressed into the glacial ice; including the atmospheric peak at 600ppm.

Figures 5&6In Figure 6 at 200 years the atmospheric concentration has dropped back to the initial condition while the firn continues to pass the remainder of the perturbation to the glacial ice.

Figures 7&8In figure 7 the glacial ice recordation is complete and in Figure 8 we see how this recordation might ideally move another 250 years further into the glacial ice.

Atmospheric concentration perturbations of differing amplitude and duration would, I think, produce similar “ideal” results.

And this is what many alarmists believe and represent (actually skeptics and alarmists alike). i.e. The ice cores are often represented (by ice-core near-perfect-recordation enthusiasts) in such a way that we are told that the atmospheric CO2 of today, at 400ppm, is at the highest level ever for 800,000 years; that the ice-core data show this is so; that the ice-cores faithfully record the peak CO2 concentration which has never been so high as 400ppm.

However, do WUWT readers see any problems with this “ideal” characterization of the ice-core recordation process? I do…and bigtime. In a future post I’ll detail my issues with this ideal portrayal but for now, can WUWT readers help me by sharing their concerns and recommendations? (Let’s not quibble too much about the “Depth Below Surface” numbers. I realize that in reality the glacial ice would likely be more compact, the spikes and gradients yet more abrupt, but have chosen my numbers for graphic convenience.)

Hints (as to exactly where the future essay is heading):

There are very substantial concentration gradients to drive diffusion processes. There are alternating seasonal temperature gradients pumping away to create activity in the system. There is an, ever present, enormous pressure gradient; giving rise to a density gradient; in turn giving rise to a diffusion mobility gradient.

The high quality recordation in Figures 3 and 4, where atmospheric CO2 is higher than what has been previously recorded, is likely quite different from the subsequent low quality recordation where atmospheric CO2 is falling*. After a hundred thermal cycles, while atmospheric CO2 is falling, just where might a thinking and reflectful engineer imagine that a great deal of that CO2 would return to?

And this is all occurring at the initial outset of ice-core recordation process (first few hundred years) notwithstanding an array of big-deal in situ and subsequent recordation distortions – each and every one of which specifically and selectively diminishes “peak CO2” recordation…selectively.

*The ice-core “Near-Perfect-Recordation” Enthusiasts are quick to point out that the firn today captures accurately the ever increasing trend observed at Mauna Loa. This is a false positive confirmation. A time will come when atmospheric CO2 is falling and that is when we will readily observe the diffusion attenuating distortion subject of this essay.

Let me say that again with different words.

A great many anomalies surround ice-core CO2 recordation of peak perturbation. Some are big (several), many are medium to small. One big anomaly (signal attenuation) comes at the outset of the recordation process (the subject of this essay). And individually each one, and more consequentially when summed collectively, all anomalies seriously diminish the CO2 peaks. And there is no known anomalous process that could have ever acted otherwise. i.e. No known process can act in such a way as to not diminish the CO2 peaks; no known anomaly can possibly make the peaks more peaked than reality. All metrological imperfections act in such a way as to diminish the recorded peaks of CO2 – and quite significantly and selectively at the peaks.

In this 200 yearlong 100% perturbation example I estimate that no more that 10% of the original signal can survive the initial attenuation of the recordation process, let alone the subsequent and substantial further attenuating distortions.

CAGW isn’t just wrong…it’s entirely and antithetically wrong. Rather than climatic poison, elevated atmospheric CO2 is the elixir of all life on Earth. And misinterpretation/misunderstanding of the ice-cores has contributed greatly to our ongoing confusion. For more info see here.


Ronald D Voisin is a retired engineer. He spent 27 years in the Semiconductor Lithography Equipment industry mostly in California’s Silicon Valley. Since retiring in 2007, he has made a hobby of studying climate change. Ron received a BSEE degree from the Univ. of Michigan – Ann Arbor in 1978 and has held various management positions at both established semiconductor equipment companies and start-ups he helped initiate. Ron has authored/co-authored 31 patent applications, 27 of which have issued.

Advertisements

581 thoughts on “An Engineer's Ice-Core Thought Experiment

  1. This is right on the mark. The time resolution of ice core slices is much better than it was even 10 years ago.
    But it’s nothing like varves in clay or loess. I read a paper that analysed a lake bed and claimed annual resolution.
    A study that analysed a peat bog claimed remarkable precision of 10 years or so in climate analysis.
    Century fluctuations would be lost in ice cores but not in varves or bogs.
    Looking forward to the follow-up post.

  2. Thanks Ronald, another angle that unambiguously shows the truth.
    What beggars belief is that these CAGW “trough dwellers” liars and rogues persist with the deceitful ways, it can only be the money – how shameful they are.

    • Robert of Ottawa
      You say

      I would think that diffusion would make the FIRN act as a kind of 40 year low pass filter

      Not merely diffusion.
      For decades I have pointed out that atmospheric pressure varies with the weather so gases would be pumped in and out of the firn by weather with resulting mixing of the gases in the firn.
      So, the gases are mixed in the firn by weather.
      This is link to one of the places on WUWT where I have made this point over past years.
      In his above article, Ronald D Voisin writes

      This glacial ice represents past seasons, nominally in layers, from 40 years to as many as 800,000 years.

      This suggests that the age of a layer provides smoothing which could be over a period as short as 40years. However, the smoothing cannot be less than the smoothing provided by the time required to seal the firn. And the IPCC says the firn typically takes 83 years to seal.
      If the IPCC is right that 83 years is the correct sealing time, then the effect of gas mixing in the firn would be similar to conduct of an 83-year running mean on data obtained from ice which sealed in less than a year.
      There is no possibility of conducting an 83 year running mean on CO2 data from Mauna Loa because that data only begins in 1958 (i.e. less than 60 years) and, therefore, the ice core data cannot be directly compared to the Mauna Loa data.
      Richard

      • Sealing times run from less than ten years up into the thousands, depending on the site, so there isn’t any typical time.
        Regarding the pumping of air through the firn, the pressure difference between high and low pressure areas runs equivalent to a thousand foot change in altitude. If there weren’t a sealed bottom, we’d be flushing a thousand feet of air through the firn every time a weather system went by, and how many times does that happen in a year? As it is, we’re compressing and expanding against the sealed layer, so lots of homogenization at the top, and progressively less as we go deeper.
        Diffusion can happen in the sealed layers, too, but very slowly. However, there are hundreds of thousands of years for that to happen. One wonders, when 120,000 years from now the apes drill down to see what the humans had for an atmosphere, whether the current 400+ CO2 spike will show up in the cores.

      • Richard, Thanx for this. If Mauna Loa comes to look something like my example, an 80+ year running mean would largely eliminate the whole of the perturbation. And Mauna Loa would look like the Future Ice-Cores.

  3. “The high quality recordation in Figures 3 and 4, where atmospheric CO2 is higher than what has been previously recorded, is likely quite different from the subsequent low quality recordation where atmospheric CO2 is falling.”
    I’m assuming that you’re holding in abeyance the reason for this asymmetry. But, if that statement was intended as logically following from the part of the essay that preceded it, could you explain a little further how you got there?

    • In a low pass filter there is no asymmetry. Deviations from the mean, positive or negative, are smoothed toward the mean. It seems this is a good model to describe the diffusion processes that will occur, lowering the peaks and filling the valleys.

  4. Thank you for bringing this up. I have been suspicious of the alleged stability of the contents of the ice cores.

  5. I was in Ann Arbor in ’78! I was a freshman, finished the BSE in Mechanical Engineering. There is no degree called BSEE from Michigan, it is called BSE in Electrical Engineering. Details…

      • Ron,
        Great stuff, it’s very nice to see the engineer’s perspective. I’m proud to be a sceptic like you, particularly as the whole basis of science is scepticism, climate science excepted.
        Knowing your background, I wonder if you worked for Applied Materials by any chance? I worked for Applied for many years at Horsham.
        Best regards,
        Chris

  6. There are obviously many uncertainties within the ice core record as is currently upheld and various physical and subsequent scientific assumptions are made to produce the palaeo measurements we all know and love. However,it is important to note that whilst the ice core science cannot readily be verified as ‘valid’ (I suppose it might in a thousand or so years time, when we could take new cores and see the current CO2 rise?) the fact that ‘changes’ in the make up of the ice are ‘measured’ it does at least give indication that actual ‘presumed’ climatic changes took place. Sure, I have big issues with the ‘calibration’ and cross correlation for these guys to produce these as past temperature/climate regimes, but the ‘deduction’ part is at least fairly logical.
    I entirely agree with the implication that the ice core record is not as ‘perfect’ as some would lead us to believe – but it IS a record, and it IS useful. Obviously, the ‘recording’ process cannot be validated as accurate (yet) but it does correlate with other palaeo datasets to a reasonable degree. Of course, there is also the cross correlation between such datasets which may thus ‘auto’ agree to some extent, but I do think that it is acceptable to consider the ice core records as at least indicative of past climatic changes.

    • Why rush to attempt to draw conclusions from measurement systems that are obviously inaccurate?
      We do have ways to measure nowadays using satellite and seaborne probes and (hopefully) unadulterated reporting. So from a purely engineering perspective, since there is no valid proof whatsoever of CAGW let the collectors collect until such time as we have trustworthy data – and throw out unreliable data where it belongs, in the garbage.

      • Not sure I follow you! The point is that the ice core ‘measurements’ show ‘something’ – of that there is no serious doubt. The systematic careful drilling and collection and recording of the ice cores does show changes within the historic ice and thus ‘something’ caused those changes. I’m not defending what they say ’caused’ the change, or how it relates to CAGW – merely that it IS there and IS detected and appears to be reflected in other palaeo datasets. The accuracy and correlation issues (including what the changes actually represent – kind of like tree rings; does increased growth mean more sun or warmth or more rain, etc, etc) still remain, of course!

    • Further…..Since there is no Human CO2 Catastrophe, why try to prove an hypothesis that is already dis-proven? Since we now have satellites and ocean probes that can correctly give us accurate data, simply continue to collect and by all means, allow interested parties to analyse it and, when sufficient samples have been taken over time, turn it into an accurate branch of science.
      In the meantime stop wasting billions of dollars on fanciful speculation.

  7. Measuring the spike at 600ppm trapped in ice after tens or hundreds of thousands of years would be a technical marvel.
    Firstly, you sample would have to be incredibly thin, not to contaminate the sample with the lower concentrations above and below.
    Secondly, who said water ice is inert? It has been known for more than 50 years that chemical reactions go forward in ice, just at a slower speed.
    Since there are regions of lower CO2 both above and below the pike, over may thousands of years I find it very difficult to believe there would not be diffusion.
    This should show up in the ice cores themselves. If they show a smooth profile of CO2, then diffusion should be suspected./ If they show a spiky profile in excess of noise, then a case could be made that diffusion was minimal.

    • Good point about ice not being chemically inert..
      And why believe that ice stays solid forever in a glacier, anyway. Any vibration (or amplified combination of vibrations) could cause patches to melt and quickly refreeze in the depths.
      And that would certainly speed up diffusion of a narrow peak… but does it happen?
      I can’t say for sure but it seems feasible and over a tens or hundreds of thousands of years “feasible” may be all that needs to be proved.
      However, Ferdinand Engelbeen knows more about this than I do and I await his input.

    • Ferd, the ice cores of Vostok and Dome C show 4 and 8 peaks of CO2 slightly after 8 peaks of temperature. The ratio between peaks and glacial lows is 8 ppmv/K over all 8 peaks. No sign of lowering the ratio for each interglacial back in time after another period of 100,000 years of migration.
      The fact that there is no flattening of the ratio is the best proof that the migration is simply negligible in the cold inland ice cores…

      • Ferdinand Engelbeen
        You say

        The fact that there is no flattening of the ratio is the best proof that the migration is simply negligible in the cold inland ice cores…

        Really!?
        Diffusion rate is driven by concentration difference. Please explain how the small differences are known to not be an approximate ‘end point’ of diffusion.
        Richard

      • Richard, migration only stops when there are no differences in concentration anymore. But for the sake of clarity, let us say that the peak value of around 300 ppmv in the ice core originally was 400 ppmv (not the 3000 ppmv of Prof. Salby) to make a comparison with the peak value today.
        That spread from the 10,000 warm years over the 100,000 cold years where nowadays 180 ppmv is measured. That means that the 180 ppmv originally was 170 ppmv + 10 ppmv received from the interglacial peak (averaged, as in reality it is a function of the distance). 180 ppmv is already at the edge of survival for C3 plants, but over land the levels are some 40 ppmv higher than in the bulk of the atmosphere, at least for part of the day.
        But that is only the start.
        The next interglacial back in time is ~110,000 years older, thus twice the time to remove part of the peak. To show the same 300 ppmv, the original peak should have been 1.33^2 higher or 530 ppmv. That is 230 ppmv to be spread over 100,000 years of cold climate or originally 157 ppmv. That is very problematic for a lot of plants, even on land…
        Finally, after 8 periods of glacial/interglacial swings, we have an original peak of 1.33^8 or about 3900 ppmv and the CO2 levels during the glacial periods then were originally negative… Even a much smaller migration, say 10% per interglacial, gives much too low CO2 levels during glacial periods.
        As most plants and animals survived the 800,000 years (and more), any huge migration of CO2 in the oldest ice cores is simply impossible.

  8. An interesting thought experiment, but it is just that — a thought experiment. It misses the mark (in my opinion) is several ways.
    1) “It peaks at 600ppm … potentially representative of prior real-world events …”
    What makes you think that such spikes could be real? What data supports the claim? What mechanism could cause the CO2 to double in such a short time? It makes for an interesting discussion, but it seems rather implausible.
    2) “And this [ideal recordation] is what many alarmists believe and represent ..”
    I haven’t seen that. What leads you to this conclusion? My understanding was that pretty much all the people in this field recognize that there is diffusion that smooths the peaks and valleys. If even wikipedia discusses this (http://en.wikipedia.org/wiki/Ice_core#Paleoatmospheric_sampling), then it is not exactly a new discovery. Sure, there are always poorly informed people in any discussion, but are there really well-reasoned papers or blog posts that don’t recognize the diffusion impacts?
    3) “In this 200 yearlong 100% perturbation example I estimate that no more that 10% of the original signal can survive … “
    Again, i would ask what experiments support this specific conclusion? I suspect that some will be coming in future posts. It would seem to make sense to start by checking the literature on the subject, rather than completely re-inventing the wheel here. I am sure there is significant research into this already. What would they say about the ice record of such a sharp spike?

    • Not sure I understand you clearly here. Let me challenge your points.
      Your point 1 is that these spikes may not exist? But we have one at the moment – unless you think that CO2 will runaway forever to Venus-esque levels. Improbable and not supported by the IPCC.
      Your point 2 is that this is all known anyway (which is hard to align with your point 1). Also, it misses the article’s key argument that the understanding of diffusion is not validated. We don’t know if we have it right. Let’s wait and see what the author follows with.
      Point 3, it’s an estimate of diffusion. There are no experiments – that’s the point. Why rely on other people’s guesses when you can make your own and understand the reasons for the estimates? Do you have any references that challenge his estimate?

      • Your point 1 is that these spikes may not exist? But we have one at the moment …
        Not really.
        1) This spike is ~ 80 PPM in the last ~ 50 years, which is about 1/2 the rate that is posited here.
        2) This spike has an easily explainable source — burning of fossil fuels. In fact, the current spike is less than the CO2 added by fossil fuels, so nature seems to be actively working to REDUCE the spike, not create one.
        “Also, it misses the article’s key argument that the understanding of diffusion is not validated. … There are no experiments – that’s the point. “
        A quick Google search reveals:
        http://www.researchgate.net/publication/228488078_Change_in_CO2_concentration_and_O2N2_ratio_in_ice_cores_due_to_molecular_diffusion
        http://www.geocraft.com/WVFossils/Reference_Docs/CO2_diffusion_in_polar_ice_2008.pdf
        http://wattsupwiththat.com/2011/01/01/antarctic-ice-cores-the-sample-rate-problem/
        http://adsabs.harvard.edu/abs/2004AGUFM.C31C..06C
        There ARE studies of this topic! There are various measurements of diffusion. Certainly there can always be new and better studies, but we shouldn’t ignore work that has been done already.
        “Why rely on other people’s guesses when you can make your own and understand the reasons for the estimates?”
        I could flip this around. This ‘thought experiment’ is indeed mostly a ‘guess’ about diffusion rates of CO2 in ice. Why put stock in this particular guess?
        I have no problem with people wanting to get a better understanding. Over all, the questions asked here are important and valid. Diffusion will certainly smooth spikes, so (as people already know), the peaks and valleys will be less extreme in the ice core data than in the original atmosphere. But just using intuition or wishful thinking about how things ought to be is not going to move the discussion forward or give specific numbers for how much smoothing there will be.

      • tjf says:
        1) This spike is ~ 80 PPM in the last ~ 50 years, which is about 1/2 the rate that is posited here.
        Are you not forgetting that global T has fluctuated by only a tiny 0.7ºC — over a century and a half?
        Just prior to the present Holocene temperatures changed by tens of whole degrees. So the current “spike” probably isn’t much of a spike at all.
        If you accept that answer, then the rest of your argument falls apart, too.

      • db, if you think that 0.7°C can produce 110 ppmv in the atmosphere in 160 years, how much more CO2 would be produced from 12°C increase between a glacial and an interglacial over a period of 5,000 years? Seems to be a little too much (~60,000 ppmv) to be true…

      • Ferdi, a 12C thermal spike from deglaciation would indeed stimulate an enormous natural source increase which would be closely followed by an enormous natural sink increase. The net CO2 spike almost certainly topped 1000ppm and likely very much more. The ice-cores then recorded a 50% spike.

      • tjf – the new OCO satellite takes issue with your #2. On land the two big areas shown on the first map release shows only moderate CO2 over the fossil fuel users in the Northern Hemisphere. The big surprise is that the large heavy clouds are either over ocean areas or over South America and southern Africa. The largest concentrations are in the Southern Hemisphere, where so many live in primitive conditions. It might be that you will have to look for this supposed spike elsewhere.

      • Ron, you forgot Henry’s law: at one side warming oceans release more CO2, but that is between 4 and 17 ppmv/K to reach a new equilibrium with the atmosphere. On the other side expanding vegetation uses more CO2. The net result is that the (deep) oceans win the battle: there is an increase of 8 ppmv/K and little change in δ13C (the latter is opposite and huge if vegetation was the cause of the increase).
        Thus a warming of 12 K will give about 100 ppmv extra in the atmosphere, not more. Be it that short living spikes are always possible, but as the ice cores show a very slow warming: 5000 years for 12 K (0.0024 K/year!) and CO2 following the temperature increase with a ~600 years lag, there is little push for a CO2 spike…

    • 1) Biologically induced CO2 spikes attend (closely follow) every temperature transition from warming to cooling. That is what is happening just now. However, Atmospheric concentration perturbations of differing amplitude and duration would, I think, produce similar “ideal” results.
      2) We are inundated with “atmospheric CO2 of today, at 400ppm, is at the highest level ever for 800,000 years”, often from people who should know better.
      3) I’ll expound on my thinking in the follow-up.

  9. but it IS a record, and it IS useful

    The question is how useful and for what. A teaspoon is useful for taking sugar with coffee, I would not dig a well with it.

    • Absolutely! – but it is no different to many proxy measurements, they have to be cross referenced with other independent data to show correlation and ‘confirm’ what they may be ‘recording’ . The usefulness depends on how accurate it is and what real time resolution can be drawn from the changes. Realistically, as this article suggests, a 200 year ‘spike’ might be difficult to ‘extract’ from a heavily compacted and condensed ‘record’, so it would be unlikely that ice core records are much use below 1000yr resolution – which for me, begs the question how do they get the oft quoted 800year time lag between temps and Co2?

      • How does that beg the question?
        It seems like it raises the question quite pertinently. What is the resolution of the ice cores?

  10. Robert of Ottawa said:

    I would think that diffusion would make the FIRN act as a kind of 40 year low pass filter.

    I think there should also be a leakage term to handle the CO_2 diffusion in ice. There is a 2008 paper on CO_2 Diffusion in polar Ice here where they say:

    …simulations for depths of 930–950m (60–70 kyr) indicate that smoothing of the CO2 record by diffusion in deep ice is comparable to smoothing in the firn.

  11. So Part II will be a deconvolution of the time-CO2 series of the ice-core data?
    With basic math as you describe, and the assumption (warmist belief) that the Antarctic/Greenland ice-core CO2 data is untarnished and without significant breaks (indicating sudden, unexplainable warming and ice loss), a de-con analysis will take you back to the atmospheric values.

  12. The process of transforming firn to ice is essentially one of compaction (ie reduction in the relative volume of the air filled voids). The excess air generated by compaction will escape upward through the overlying younger layers in the firn column as long as there remains some connectivity between the voids. This flushing effect would tend to reduce the peak CO2 concentration at least during the first 40 years of a period of increasing atmospheric CO2 concentration.

    • It’s much worse than that. Every time the atmospheric pressure changes (as it does constantly) air gets “pumped” in and out of the snow and firn, and this propagates downwards, though of course less and less as depth increases. By the time the firn turns into ice, say after 500 years (it often takes longer), the air is a mixture with an age range of at least 500 years.
      And there is at least one more complication. Occasionally an exceptional thaw may melt the snow on top of an icecap and create a more or less impermeable ice crust. This seals off the snow below the crust, but at least at first intensifies the “pumping” effect on top of it. This however mostly affects Greenland during interglacial periods, it is very rarely or never warm enough for large-scale thawing even in coastal areas of Antarctica.

      • tty and fhsiv,
        You guys are exactly right. Several commenters have pointed to diffusion studies of the ice. What I’m pointing to is the attenuating diffusion that occurs before ice in the firn.

      • I am really cunfused here how would co2 get trapped during a warming phase anyway. If we go through a prolonged warming phase how much ice would becreated out of the firn anyway. It seams to me very little witch means the layers you are locking at would be very thin and all issue stated above would be greatly magnified. If a rise in atmospheric co2 is cause by a prolonged rise in temperature then the record of this in the ice cores might not exist at all.

    • fhsiv, you forget that the migration has 40 years of time to equalize the levels. That makes that the levels at bubble closing depth (72 m) are hardly older than in the atmosphere and as Etheridge measured, the same CO2 levels are found in still open pores and fully closed bubbles at the same transition depth.

  13. This is article is just conjecture. But taken with the fact that CO2 proxies like plant stoma indicate that CO2 levels were higher than present day about 13,000 years ago add teeth to such an analysis. Why would anyone assume the ice core proxy of CO2 is correct and ignore the leaf stoma proxy? Science would not do that. Political Science would.

  14. I like your writing style, Ronald; a bit verbose, but very clear. Wonderful graphics!
    Your “hints” seem more like spoilers to me, but then, I’m a retired engineer. I look forward to part 2, and to finding out whether I’ve correctly anticipated how you’ll modify these diagrams to depict the confounding factors you mentioned.

  15. In part 2, it might be instructive to depict how a thin dust layer from a big volcanic eruption gets recorded in the glacier, and to address why it would not get vertically dispersed as we should expect of a gas.

    • Star Craver, Thanx. Probably won’t get into dust layers, but obviously diffusion is a function of mobility and dust isn’t generally mobile; except in the cases of melt – pretty much confined to the dimensions of the melt.

  16. “What makes you think that such spikes could be real? What data supports the claim? What mechanism could cause the CO2 to double in such a short time? It makes for an interesting discussion, but it seems rather implausible.”
    Actually there are at least two potential mechanisms for such a spike, an ocean flushing event, i. e. a turnover of a stratified ocean and a major cosmic impact (particularly of a comet and/or if it happens in a carbonate basin).
    And for those rushing out to defend the reliability of the ice-core record: It is accepted by everybody familiar with ice-core data that such brief events as postulated here are more or less completely smoothed out. They might be faintly discernible in the highest-accumulation cores (e. g. WAIS Divide) if they happened in the recent past, but definitely not deep in the past in the long low-accumulation cores like Vostok or Dome C.

      • Quoting David Middleton:

        Ice cores from Greenland are rarely used in CO2 reconstructions. The maximum usable Greenland record only dates as far back as ~130,000 years ago (Eemian/Sangamonian); the deeper ice has been deformed. The Greenland ice cores do tend to have a higher resolution than the Antarctic cores because there is a higher snow accumulation rate in Greenland. Funny thing about the Greenland cores: They show much higher CO2 levels (330-350 ppmv) during Holocene warm periods and Pleistocene interstadials. The Dye 3 ice core shows an average CO2 level of 331 ppmv (+/-17) during the Preboreal Oscillation (~11,500 years ago). These higher CO2 levels have been explained away as being the result of in situ chemical reactions (Anklin et al., 1997).

      • Greenland ice cores are not reliable for CO2 levels, as they have not only more sea salt (including carbonate) deposits than (inland) Antarctic cores, but also frequent highly acidic volcanic dust from the nearby Icelandic volcanoes. The results is in-situ extra CO2 formation, increasing with the (now abandoned) wet measurement method where all ice was melted, showing increasing CO2 levels over time…
        Nowadays, they use cold grating of the ice under vacuum over a cold trap (to catch water vapor) or full sublimation of all ice over a cryogenic trap and subsequent separating of all components, including the different isotopes over a mass spectrometer.

      • I have no direct knowledge of Icelandic volcanoes in other parts of the Holocene, but I have read somewhere that they test for the acidity of the snow layers and have frequent parts which show high acidity, dust and CO2 levels. If they use the old method of melting all ice and removing CO2 under vacuum, the CO2 levels start high and get coming with time, which is caused by the reactions of sea salt carbonates with volcanic dust… Icelandic volcanoes all spew deep magma lava, ash and gases, which includes SO2, HCl and HF. The latter is quite toxic for animals (and humans): about 60% of all grazing stock died in Iceland during the Laki eruption of 1783-1784 and 22% of the population died of famine…

  17. I’m waiting for Ferdinand Engelbeen to comment, too.
    When I first read about how ice core CO2 was measured, I was as sceptical as engineer Ron. But I have, as a trusting sort of guy, been happy to accept the ice core specialists’ reassurances that their record is a true picture, and I think Ferdinand supports that.
    Nevertheless I have this nagging worry that 800,000 years have passed during which atmospheric CO2 concentration has not, according to the core work, deviated by much more than a few percent: seems pretty unlikely, considering its vast and many-faceted exchange with a changing biosphere. To a biologist, that just doesn’t ring true. Maybe Jim Steele will give his opinion as well: always worth listening to.

      • Interesting to note that it is dust, not CO2 that peaks just before periods of rapid temperature increase.

      • MikeB,
        If you look closely you will see that CO2 is not rising and falling “in line with temperature”, but rather, that CO2 follows temperature.
        The time scale makes that difficult to see on that graph, but it is easy to see on shorter time scales — none of which has ever shown that ∆CO2 causes ∆T.

      • Erik, the dust does not line up at all with temperature. Notice the time scale is 100s of thousands of years.

      • Pretty fair counter, Mike
        Should teach me to get out of the armchair and pull up the data before commenting. Probably a false memory syndrome on my part. The graph my retrieval system pulled up has the whole record departing minimally from 250-260ppm, clearly that’s not the current picture. I certainly had not absorbed that the values at the peak glaciations went down below 200 ppm.
        The consistency with which the shape tracks the glaciations also lends authority to the ice core data.
        Comment withdrawn.

      • The dust deposits are mainly in the coldest periods of the earth: less water vapor, less clouds, less rain and more wind which blows desert dust up the heights of the Antarctic inland ice…

      • Barry, Ferdinand,
        I am aware of the timescales and that the peak dust coincides with cold, dry climates. A WAG on my part is that enough dust may accumulate in the ice to lower the albedo to the point to cause rapid ice melting. I’m also open to the dust levels dropping dramatically due to increasing precipitation brought on by warming.

    • To a biologist, that just doesn’t ring true.

      HA, …. more often than not it takes a biologist to get “sense n’ sensibility” back into a scientific discussion.

  18. My information comes from communication with several ice core experts, including many hours with
    Fritz Koerner, one of the few people to core in the Arctic and Antarctic.
    http://www.canada.com/ottawacitizen/news/story.html?id=b618ecb7-9973-4c32-a60d-60c689a60177
    The Antarctic ice core record (Petit et al 1991) was quickly seized on as it appeared to show that CO2 was driving temperature. Koerner told me at that time that his Arctic cores were showing temperature preceding CO2.
    1. It is also a 70 year filter depending on which glacier you use.
    2. Since it takes between 40 and 70 years for the bubble to be enclosed, which year does it represent?
    3. They then apply a 70 year smoothing average to the data.
    4. A 2000 year stomata record and the 19th century instrumentally measured direct air samples bot show much higher atmospheric levels and much higher variability.
    4. In some locations on the ice, years of data is obliterated by wind removing both recent snowfall and then granular snow abrading previous layers before they become ice. The wind blown snow piles up in other areas creating a false annual layer.
    5. There is so much movement of water through the ice, that severe contamination of any bubble is inevitable.
    6. As the ice becomes plastic it deforms and the bubbles are squeezed out.
    7. The ice generally becomes plastic at approximately 50 m, but at a certain depth the plasticity eliminates the ability to determine annual layers.
    7. I understand that at 2000m it takes 8m of ice to get a single gas sample. At that depth, 2000m represents at least 10,000 years of deposition.
    8. Even if you accept all the limitations, the record shows temperature changing before CO2 in absolute contradiction to the fundamental assumption of the AGW hypothesis.
    9. Jaworowski gave testimony to the US Congress
    http://www.mitosyfraudes.org/Calen5/JawoCO2-Eng.html
    and identified the basic problems with ice cores based on 40 years experience.
    http://www.warwickhughes.com/icecore/zjmar07.pdf
    It is why he was so viciously attacked. Today, that is a measure of how close to the truth and therefore how big a threat to the political agenda a scientists has become.
    10. As Richard Benedick, a deputy assistant secretary of state for the U.S. State Department, said: “A global warming treaty must be implemented even if there is no scientific evidence to back the [enhanced] greenhouse effect.”
    11. And former Senator Timothy Wirth, who orchestrated the appearance of James Hansen at the political charade
    http://www.pbs.org/wgbh/pages/frontline/hotpolitics/interviews/wirth.html
    at which the entire issue went global said in 1993, two years after the ice core record appeared, “We’ve got to ride the global warming issue. Even if the theory of global warming is wrong, we will be doing the right thing …”
    12. Of course, Hansen was subsequently replaced by Gavin Schmidt, who’s performance with the 2014 fiasco indicates it is business as usual.

    • There was also this.
      And this:
      Ottmar Edenhofer, Co-Chair, IPCC WG-3:
      “…one must say clearly that we redistribute de facto the world’s wealth by climate policy. One has to free oneself from the illusion that international climate policy is environmental policy. This has almost nothing to do with environmental policy anymore…”

    • Tim,
      Besides the fact that most in the climate community nowadays accept that temperature changes precede CO2 changes in the ice record (except Al Gore and a few alikes):
      1. The “smoothing” of the CO2 record is between 10 and 600 years, depending of the accumulation rate.
      2. The average CO2 age can be calculated via a firn densification model, depends of 1.
      3. Not that I know. References?
      4. Stomata data and most historic measurements were taken over land: high variability and too high levels. Interesting for resolution and probable causes of variability but unreliable for absolute levels.
      4′. Not important, only shifts the ice age – gas age difference which is anyway unsure for low resolution ice cores.
      5. Movements of water in ice at -40°C? Seems rather difficult to me. Ever tried to skate on ice of -20°C?
      6. Bubbles squeezed out to where? To the atmosphere? The bubbles simply follow the ice stream.
      7. Again not important except for determining the ice age – gas age difference.
      7′. 10 m of ice for one sample? maybe 20 years ago, now they used for Dome C: “At the Bern laboratory, four to six samples of approximately 8 grams from each depth level (0.55m intervals)…”
      8. Agreed.
      9. As far as I know, Jaworowski wrote a letter, but never testified for congress. And as said elsewhere, his knowledge of ice cores was never updated after 1992 and certainly wrong on such elementary points that I personally can’t believe anything that he wrote after 1992.
      10-12 That indeed is politics, not science…

    • Bulls eye. Dr t.
      ” how big a threat to the political agenda a scientists has become”. My childhood nickname was bull’. Lol.

  19. Your point is well taken, BUT, note well that the absence of peaks today does not imply the presence of peaks in the past. It simply establishes that we could not tell if there were.
    The real point is that one cannot rely on the ice core record to give anything more than what it gives — a very coarse grained average picture of past CO_2, subject to distortion by various local processes that we can no longer detect or correct for.
    I disagree, BTW, about the impossibility of creating at least mild positive anomalies relative to the average and indeed suspect that some of the Laws data illustrates exactly that. Remember, the modulation of CO_2 across layers is dependent on local conditions, which may not be homogeneous. Some years the CO_2 record may be more pristine than others because e.g. it was unusually warm, or cold, or snowy in just the right pattern.
    rgb

    • There are several pCO2 paleoproxies including stomata amd alkenones. None have high resolution as this post is discussing. But rough past delta CO2 levels from whatever method (ice cores, paleoproxies) lag delta T on the order of at least 400 and probably about 800 years. Simply a comsequence of Earth’s surface being mostly ocean, Henry’s Law, and Le Chatellier’s Principle. Essay Cause and Effect in ebook Blowing Smoke covers this. It debunks several recent warmunist efforts to show the opposite. Embarassing attempts to overturn established physical chemistry.

      • Stomata have very high resolution (one year). The difficulty is to date the leaves and find homogenous series of fossil leaves. However there are a number of Late Glacial and Holocene series, and one from the last interglacial (MIS 5e), and they all indicate considerably more short-term variation i pCO2 than ice-core data.
        Also the fact that ice-core data from high accumulation sites, such as WAIS divide show more variability than low-accumulation cores is certainly suggestive.

      • tty, the main problem with stomata data is that they are influenced by local CO2 levels, not only “background”. One of the main places for stomata history is in the SE Netherlands, where considerable changes in land uses occurred in the past 1000 years: from sea to land, from marshes to agriculture and forests (for mining) and not at least industrialization in the past 150 years, all in the main wind direction. Even the current main wind direction (mainly SW, second NW) may have changed during the Little Ice Age, where at least winters were severe with blocking highs and wind from mainland Eastern Europe…

      • tty, the main problem with stomata data is that they are influenced by local CO2 levels, not only “background”

        Isn’t every proxy related to local inputs? Professor Xavier may be able to know at a distance but I can’t see how he does it. \so how would ice cores.
        If the argument is that stomata pick up local inputs because they aren’t smoothed by averaging over time like the ice cores then are… isn’t that arguing for poor resolution/less meaning of the ice cores?

      • MCourtney,
        The main difference is that in the bulk of the atmosphere, that is everywhere from the North Pole to the South Pole over the oceans and (ice) deserts or above a few hundred meters over land, one can find the same CO2 levels within +/- 2% of full scale. Even including the seasonal changes and the human contribution.
        In the first few hundred meters over land, one can find 550 ppmv at night under inversion and 280 ppmv during a sunny day. The historical measurements at Giessen (SW Germany), the longest historical series, show a variability of +/- 68 ppmv (1 sigma), those at Mauna Loa +/- 4 ppmv and the South Pole +/- 2 ppmv…
        That wouldn’t be a problem if the local bias was constant and the local variability was related to a global variability, but for both, there is little/no historical information and the information from modern stations at the same places as the historical one’s is not very hopeful…

      • MCourtney January 25, 2015 at 3:09 pm

        If the argument is that stomata pick up local inputs because they aren’t smoothed by averaging over time like the ice cores then are

        They will claim most anything to discredit the stomata proxies.
        All plant growth for a particular locale is determined by “average rainfall”, …. not a once-in-a-while thunderstorm or dry spell.
        And likewise, the size and/or number of stomata is in fact the result of a per se “CO2 ppm averaging”. A leaf producing plant is smart enough not to over-produce or under-produce the stomata required for optimum plant growth. And “Yes”, …. “smart” like an Acacia tree is smart. http://www.newscientist.com/article/mg12717361.200-antelope-activate-the-acacias-alarm-system.html

      • Ferdinand Engelbeen January 25, 2015 at 4:18 pm

        In the first few hundred meters over land, one can find 550 ppmv at night under inversion and 280 ppmv during a sunny day.

        YUP, but all leaf producing plants are paid-up members of the United Nature Growers Union #169, …… they don’t work at nighttime.

    • Exactly, Dr.! The data is pre-smoothed and smoothed in ways we cannot predict. It is useful on large time scales only, and we cannot really compare measurements we get today with it. It’s very plain to me, the first time I devoted some thinking to this subject I came to this conclusion. Thank you.

      • Richard,
        The “displacement” as said by the late Jaworowski is a real mistake of him, which is based on the assumption that there is no difference between ice age and gas age in ice cores.
        Everybody with the slightest knowledge of ice cores will know that the average gas age in the enclosed bubbles is (much) younger than the surrounding ice, for the simple reason that the pores in the firn remain open for many years until a depth of ~72 m and more. For Law Dome that means 40 yearly ice layers. The average gas age in the transition zone then is ~10 years older than in the atmosphere, that is 30 years younger than the surrounding ice. That difference remains there for as long as the ice exists.
        The Law Dome results were published in 1996 by Etheridge e.a., together with a lot of answers to other objections of Jaworowski. He didn’t change his mind and still accused other scientists of “cheating”, while it was he who was wrong. Not only on that point, but on a lot of other points too.
        Thus sorry, he might have been a specialist in radiation dangers and metal (ion) migration in ice cores, but his knowledge of CO2 in ice cores was completely outdated and completely wrong.

  20. Quote: “Rather than climatic poison, elevated atmospheric CO2 is the elixir of all life on Earth.”
    No matter if the ice core record of CO2 is reliable or not, the suppression of the actually obvious fact above by climate alarmism is maybe the greatest travesty of the current propaganda battle. The vilification of “The Gas of Life” has gone so far that many people don’t even know the vital role of CO2 for the biosphere on Earth. And schools do brainwash our kids so absurdly biased against “evil” CO2 that many pupils think CO2 is the most abundant gas in the atmosphere and has no positive effect at all…
    What a sad and insane state of minds after 250 years of enlightenment !

  21. I have never understood why anyone places any faith whatsoever in proxy measurements. You have no way of knowing how to correlate to past temperatures. And correlation between proxies is to be expected…after all how do you think the proxy methodologies are developed in the first place. You keep tweaking your methodology until you get a good correlation to accepted proxies.

    • I am not sure how that relates to CO2 in this case: the ClimateAudit article is about the 18O/16O changes in the water molecules of the ice. That ratio is a proxy for the ocean temperature at the place of origin of the water vapor that is deposited as snow. CO2 on the other hand is evenly distributed over whole Antarctica and shows very little difference between ice cores for the same average gas age.
      The main problem is calculating the average gas age, as that depends of the snow accumulation rate and temperature, which of course is different during glacial periods than during interglacials…

      • Ferdinand, neither Ellen’s, nor Lonnie’s data are available for scrutiny, so I don’t know what they contain. Do you? I was called a liar some years ago by an Antarctic researcher for making this claim. If you do have access, where is it archived? A rather obvious assumption to make is that there’s something worth hiding.

      • I was reacting to the fact that the Thompsons only show the ocean temperature proxy record, which differs from core to core, depending of the catch area of the water vapor. Not about CO2 levels.
        But I agree with Steve McIntyre that they should have put all their data into a public file…

      • It may be too late to get your attention, Ferdinand, but I’ve been looking at the ice core data pretty carefully to see how it might affect the simplest possible log model between CO_2 and temperature, unlagged, unsmoothed. The problem is this — there isn’t particularly good agreement between Laws, Siple, and Mauna Loa over the last 165 years. See this plot (directly from data from the relevant sites):
        http://www.phy.duke.edu/~rgb/cCO2oft.jpg
        Here are the problems I see in this:
        a) Laws goes up, Siple goes down, and vice versa. Several times.
        b) Laws goes down (or remains nearly flat) for order of decades. Say what? 1850 to 1870, 1895-1905, 1940-1950. 1940-1950? Again, say what? This was probably the fastest expansion in industrial capacity the world had ever seen. It was also the end of the very rapid rise in global temperature through the 30’s and early 40’s. Siple goes up, sharply across exactly this same period.
        c) The blue line is a reasonable smooth fit to the post 1958 ML data that intersects the nearly common Laws/Siple numbers in 1950. It is “reasonable” because the ML data is, in fact, remarkably smooth with only tiny deviations from this curve, and indeed it appears to regress to this curve when it deviates a tiny bit from it in e.g. the 1980-1990 time frame. Laws is even smoother. The three (ML, Siple, Laws) all appear to nearly perfectly merge in maybe 1955 to turn into this nice, smooth, rapidly rising curve.
        But before 1955 the Laws/Siple data is anything but smooth. It isn’t even monotonic. In Laws, CO_2 appears to (if anything) be rising linearly from 1870 to 1940, punctuated on either end by drops that as far as I can think of have no rational explanation within the confines of the current beliefs about climate and CO_2 and atmospheric chemistry. Siple, OTOH, has (obviously) different resolution but still manages a sudden, sharp drop just after the turn of the 20th century and has a trend indistinguishable from linear with this number of data points over the interval.
        I freely confess, I can make absolutely nothing out of this. It makes no sense. It is rather difficult to believe that in 1955 the entire character of atmospheric chemistry changed after 100 years of change according to an entirely different rule. It is also difficult to believe that either Laws or Siple is recording a well-mixed atmosphere that can fairly be compared to Mauna Loa. Finally, because of the profound change in character of Mauna Loa direct measurement and Laws inferred CO_2 at almost precisely the beginning of Mauna Loa’s existence, we have two serious problems to contend with. One is that we absolutely cannot extrapolate backwards and assume that Laws is an accurate proxy of what Mauna Loa would have read. Something changed and we do not know what it was, but it completely suppressed the noise and variability clearly visible in the pre-Mauna Loa Laws data, meaning that the proxy itself is enormously suspect, doubly so given the systematic and serious disagreementts with Siple. The variances make almost a discontinuous change at precisely the point where the series all come together at ML.
        That makes one wonder about the second possibility — that there are thumbs on the scales, somewhere, somehow. Did the people analyzing Laws “know” the Mauna Loa result? Almost certainly. Did they have to accept or reject samples as “good” or “bad” for inclusion in their analysis? Almost certainly. Would it have been possible for them to use their knowledge of “the right answer” — consciously or unconsciously — to select the samples for inclusion so that they were in good agreement (since the virtue of their entire body of research depends on this agreement — if it is poor what is the point)? Well, yeah, it could have happened.
        One hopes that the researchers were rigorous and pristine in their approach and used precisely the same procedures for acceptance and rejection of samples (or just used random numbers, always a good idea in statistical analysis) to obtain the Laws/Siple series. Unfortunately, the sharp change in the variance in the data at precisely the junction between the three series strongly suggests otherwise. Either that or else there is some sort of natural process that from time to time scours 5 to 10 ppm of CO_2 out of the atmosphere over a single decade in the teeth of what surely was a steadily increasing anthropogenic contribution, or there is an occult natural process with substantial decadal variability that is responsible for decadal swings in the rate of CO_2 accumulation either way, that by pure chance became a consistent producer of CO_2 at an entirely new rate in 1955 or thereabouts after varying substantially for the 100 years before around a substantially lower rate.
        Or, we could conclude that the ice core data are simply not very reliable after around 40 years. That is actually one of the best explanations. Laws might well be accurate for around 30-40 years before the present (the present of the collection of the cores). Beyond that, the data might simply be showing that there is some sort of process — quite possibly associated with a pressure driven phase transition of sorts, as that would explain the change in both variance and slope — that completely alters the ice and renders the same “fit” that correctly gives Mauna Loa post 1955 useless for interpreting samples 50 or more years old.
        rgb

      • Robert (rgbatduke), always great to hear from you. You ask why the ice core data doesn’t fit your smooth curve. First, let’s look at all of the ice core data that covers the period you discuss, 1850-present. There are five analyses covering four ice cores during that period, viz:
        https://wattsupwiththat.files.wordpress.com/2015/01/mauna-loa-and-ice-core-data-1850-2010.jpg
        Your graph is:
        http://www.phy.duke.edu/~rgb/cCO2oft.jpg
        and you ask why the ice cores don’t correspond to the blue line, which is your hypothetical CO2 curve.
        The reason is simple—the CO2 emissions don’t follow the blue curve, so there is no reason to think that the atmospheric CO2 would follow it either. Here is the actual curve of the CO2 emissions over the period:
        https://wattsupwiththat.files.wordpress.com/2015/01/global-emissions-gt-c.jpg
        As you can see, far from following your smooth curve, the emissions actually agree with the ice core data quite well, including a relatively flat spot from about 1935-1950. Presumably this reflects the effect of the Great Depression on global industry. You report being surprised that the ice core CO2 data goes up fairly linearly from 1850 to about 1950 … but in fact, that’s exactly what the CO2 emissions did. Not only that, but the fit is even better once you include the exponential decay of the emissions.
        Finally, I don’t find the small differences between the ice core results to be particularly surprising. They all seem to fall well within what I would expect from observational data.
        Best regards,
        w.

      • “As you can see, far from following your smooth curve, the emissions actually agree with the ice core data quite well, including a relatively flat spot from about 1935-1950. “
        Apples and oranges. One is a total accumulation of changes, and the other is changes. Accumulate the latter, and the “flat spot” disappears.

  22. Reblogged this on gottadobetterthanthis and commented:

    As an engineer with a thorough education and understanding of the physics of matter, especially in the solid state, I take it as certain that all variation and every gradient has forces acting to reduce and eventually eliminate the variation and gradient. There are mechanisms active and available to facilitate in all circumstances. The question boils down to time. The rate of the action is not always clear or quantifiable.
    It seems to me the actual rates involved in ice cores are slow enough to give us a reasonable picture of the past, but it is most certainly attenuated. Keep in mind that the averaging is not “average” but forcing toward zero. The farther from zero at any given point, the greater the tendency toward zero at that point. Peaks spread and dissipate. They don’t just average. The physical actions working in the ice, on the bubbles, et al., are not a smoothing so much as a squashing toward the lowest possible uniform value. Not that the natural amount of CO2 in ice is zero. That is, ice is made of H2O. CO2 will eventually diffuse out of the ice where there is no boundary condition imposing an external limit or partial pressure

    • Lonnie, there is attenuation in the firn zone, as there is a lot of mixing time and migration in still open pores, be it that the migration speed reduces with depth as the pores get smaller under pressure.
      There is no measurable migration in the extreme cold central highland cores and certainly not toward zero, as the minimum levels found still are 180 ppmv. Thus if there was migration, that will level of the (~300 ppmv) peaks and distribute that over the 180 ppmv, not zero.
      Thus peaks may spread (but there is no sign of that), but CO2 doesn’t disappear in the ice, the ice matrix is too small for CO2 (we are not talking about metal -ions- in the ice matrix). Put a Coke bottle in a freezer: all CO2 comes out of the ice and the bottle may explode if not strong enough.

  23. @ Ronald D Voisin

    In this 200 yearlong 100% perturbation example I estimate that no more that 10% of the original signal can survive the initial attenuation of the recordation process, let alone the subsequent and substantial further attenuating distortions.

    I’se just luv “original thinkers” who employ their nurtured “common sense, logical reasoning and/or intelligent deduction” personality traits …. for resolving questions and/or questionable claims about the natural world around us.
    The “gist” of the above quoted comment pretty much exactly agrees with my “thoughts” on the subject in question. Thoughts that I have been touting for the past several years in opposition to the claimed “accuracy” of atmospheric CO2 ppm being determined via Ice Core proxy data.
    Especially the Greenland Ice Core proxy data because I do not understand how it is possible to assign a “date” to the yearly accumulation of glacial ice ….. given the fact of extreme melting at the onset of the current Interglacial (21,000 BP) and/or during the Holocene Optimum (8,000 – 4,300 BP), … Minoan WP (3,500 – 3,200 BP), … Roman WP (2,300 – 2,000 BP), … and/or Medieval WP (1,200 – 900 BP).
    Fifty (50) years of rapid melting could easily “wipe-out” 500 to 1,000 years of glacial ice accumulations.

    • Samuel, the Greenland ice core record was from the summit, where besides a few re-melt layers there was little melting over the past 110,000 years. The relative high accumulation rate was sufficient to count the full 110,000 yearly layers until near the bottom where the ice was mixed/melted.
      There are differences in density between summer and winter snow, which can literally be seen by eye even in de ice and can be measured by light diffraction, conductivity, radar, echo,… when the layers are compressed with depth.

      • the Greenland ice core record was from the summit,

        Ferdinand, a couple quick facts for you.
        1) Greenland’s highest mountain – Gunnbjørn Fjeld, Island of Greenland, 12,119 ft (above sea level)
        2) The (glacial ice) thickness is generally more than 2 km (1.2 mi) and over 3 km (1.9 mi) at its thickest point
        So Ferdinand, are you telling me that the 10,032 feet of glacial ice sits on top of that 12,119 feet mountain summit ……. for a total height of 22,151 feet in elevation?
        And Ferdinand, just what kept the winds from blowing most of the snow off the summit?

        • Samuel C Cogar, challenging Ferdinand’s geography

          So Ferdinand, are you telling me that the 10,032 feet of glacial ice sits on top of that 12,119 feet mountain summit ……. for a total height of 22,151 feet in elevation?

          Absolutely not. Your impression is backwards.
          If all of the ice were removed, Greenland would be a very wide basin. Almost like the North American continent, but shorter across from the east coast to west coast. The mountains begin right at the coastline, they rise very sharply to a peak about 50-75 km inland. This is the location of the mountain peak you mention – right next to the coast! The row of mountain peaks loops completely around the central basin – think of the Appalachian mountains dropping down to the Mississippi, Ohio, Tennessee and Missouri River basin in the center, then rising back towards the Rocky Mountains in the west. But there is then a very sharp drop right back to the west coast. No Great Basin, no second or thrid set of mountains like the Sierra Nevada and coast CA-OR-WA mountains and volcanoes.
          That central basin in Greenland has been pressed down so low by the weight of glacier ice above it that the “rock” is below sea level. If the oceans could reach it when the ice were removed, the middle of Greenland – where you seem to think there are mountains by the tone of your question – is underwater.
          So: East coast, rocky and glacier-covered mountains for 75 km, a north-south bare rocky peaks with little or now snow cover of 2-4,000 meter tops with thousands of passes and low spots between with some ice cover, rocky slopes sloping down to the center basin of Greenland that trap the 100,000 years of ice in the middle, a rising center mass of ice that is 3200+ meters thick at the middle over a rock bottom that is a few hundred meters under sea level, then a rising slope of mountains and glaciers trying to flow to the middle (but can’t!), then a second rocky series of bare mountain peaks and high passes, then a sharp, short decline of rocky slopes and glaciers facing west down to the coast.

      • Samuel, RAC has answered your main question, only one addition:
        Most of the winds at the Greenland (and some parts of Antarctica) ice sheet are katabatic: they blow from the summit to the outer regions. At the center there may be hardly any wind (not always, but most of the time).
        Some people here did transverse the Greenland ice sheet East to West and used the katabatic winds to sail down once they passed the highest point…

  24. The big problem I see with the presentation of the “recordation” is the shape of the CO2 pulse.
    Modeled as symmetrical. It seems arguable that any fast CO2 release rate (50-100 years) (anthropogenic, vulcanism, marine clathrate release, bio-methane sourced) would then have a much slower, (order of magnitude) removal rate by the CO2 sinks (ocean solution, biological &, geochemical sequestration). Thus, the recorded signal shape would not be symmetrical, and the high values would have a much longer imprint time on the accumulated ice layer than as shown.

    • Further, the shape of the decay could arguably be modeled as a logarithmic half-life decay, i.e. not a linear removal.

  25. One problem with this article is that the atmospheric lifetime of CO2 is centuries. Individual molecules have a short lifetime (~ years), but once they leave the atmosphere they continue exchanging with CO2 in the oceans, for centuries.

      • That is not always the case: a lot of CO2 sources are rapid one-way: volcanoes, burning vegetation,… The reverse time needed to remove the extra CO2 may be much longer: regrowth of vegetation is quite rapid, but uptake of CO2 by the oceans is slow (fast for the surface, but that has a limited capacity). An extra CO2 uptake by vegetation once the forests are mature also is very slow. Other sinks like rock weathering even much slower…
        The current human emissions are ~10 GtC as CO2. With the 110 ppmv extra in the atmosphere, the increased pressure puts ~1 GtC extra in vegetation, ~0.5 GtC in the ocean surface and ~3 GtC in the deep oceans. That makes that the e-fold decay rate for the extra CO2 above equilibrium is ~50 years.

  26. @Ronald Voisin,
    Got to simmer on this first, but offhand the firn layer is some 80-100 meters thick and it’s the accumulation rate which determines the ice age at which the firn gets solid at those depths. On Greenland that may be around 50-100 years, on the highest domes of Antarctica, where the long CO2 records come from (Vostok, Dome C / EPICA), you think more in several thousands of years. This difference is reflected in the gas age / ice age chronology’s. See http://www.clim-past.net/3/485/2007/cp-3-485-2007.pdf and
    ftp://ftp.ncdc.noaa.gov/pub/data/paleo/icecore/antarctica/epica_domec/edc3-timescale.txt

  27. First, Ronald, thank you for your interesting article and graphics. And in general, the point you make is correct—the ice cannot increase the amplitude of up-down swings, it can only reduce them.
    With that said, however, the real issue is “how much”. Here, for example, is some real-world data:
    http://wattsupwiththat.files.wordpress.com/2010/06/mauna_loa_ice_core_co2_1000_2010.jpg
    Note that the ice cores have captured the recent increase in CO2 with good fidelity.
    So while your point is correct, in general that it is a difference that makes little difference. In part this is because some ice cores have quite short firn-closure times. Greenland ice cores are an example of this.
    The second reason that it makes little difference is that you show a very slow slope down through the firn, with the firn essentially capturing the instantaneous CO2 concentration and holding it until it is frozen into the ice.
    But this is NOT at all what happens. Instead, the firn constantly equilibrates with the current CO2 levels through diffusion. The difference between the firn and the ice is that the firn is NOT closed off from the upper atmosphere. The firn doesn’t have sealed bubbles as the ice does. It has open cells that connect to the surface. As a result, the line you have drawn representing CO2 concentrations in the firn is incorrect. It should be nearly vertical. It is only when the bubbles finally close at the bottom of the firn where it turns into ice that the CO2 stops equilibrating with the surface.
    The main effect of this is to make the age of the ice different from the age of the air enclosed in the bubbles in the ice. When the bubbles close on e.g thirty-year-old snow at the bottom of the firn, because of the constant diffusion they capture air that is little different from that of today. Yes, if the up-and-down of CO2 were rapid it would not be captured accurately … but even in modern times we’re only talking about a change of a percent or so per year, and that’s only in one direction. For that kind of slow change, the ice seems to capture the variation with good fidelity.
    Best regards,
    w.

    • As I understand the problem Willis, ice core data of the last 60 years are so will not be reliable regards to CO2 because of the diffusion problem, hence your Mona chart is suspect.

      • mpainter, the diffusion in firn has plenty of time: 40 years to equilibrate with the atmosphere, less when the pores are more and more closing near the bottom.
        Migration in the ice itself when the bubbles are fully closed is very small in “warm” coastal ice cores and only broadens the resolution, in much colder inland ice cores not measurable at all.

      • What is your point Ferdinand?
        Are you trying to say that the referred data (Mauna Loa chart) is reliable?
        That the firn closed very quickly?

      • mpainter, see my more complete response below: the difference in air composition at the bottom of the firn is only 7 years with the atmosphere. Thus the ice core follows the atmosphere very closely, because of the diffusion in the firn. The largest difference is caused by the different depths where the first and last bubbles are closed…

    • Willis, the good fidelity you speak of can only exist while the atmospheric CO2 is higher than what has been previously recorded. As soon as it reverses, fidelity will be hugely compromised.

      • Thanks, Ron, but your attempt at science by assertion doesn’t go far enough. When the cycle reverses, as you point out, the ice core records will reduce any peak … but by how much?
        As Ferdinand points out, the averaging of the CO2 values that goes on in the Siple cores ranges from perhaps twenty to forty years … so as long as the reversal takes significantly longer than that, fidelity will only be very marginally compromised, not hugely.
        w.

      • Samuel, that depends of the average temperature, thus most will be seen in coastal ice core at lower elevations. Melt layers are easily detected in an ice core and do influence the ice age – gas age difference and the smoothing of that part of the ice. That can be calculated.
        In the Siple Dome ice core they had one melt layer over a period of 90 years. At the inland Vostok and Dome C ice cores I haven’t heard of any melt layers.
        Snow blowing away is not a real problem: it influences the dating with missing years or even decades, but that is already coarse even without such events. Some cross dating is done if dust is disposed from huge historical volcanic explosions, but still problematic.

      • Stupid mistake from the late Prof. Jaworowski: he looked at the column of the ice age in the table of Neftel for the Siple ice core. But CO2 is not in the ice, it is in the gas bubbles, which are by definition younger than the surrounding ice, as CO2 can freely migrate in and out over the 40 and more years that the ice layers are stapling. Thus the average gas age is much younger than the ice age. If you take the right column in Neftel’s table, there is no “arbitrary correction” at all.
        See: http://www.ferdinand-engelbeen.be/klimaat/jaworowski.html

      • @Ferdinand Engelbeen: January 27, 2015 at 4:42 pm

        If you don’t (want to) understand that adding 4 ppmv extra CO2 per year to a system which is more or less in equilibrium (with a small variability of not more than +/- 1 ppmv/year)

        Shur nuff, Ferdinand, and the air pressure where you reside is more or less in equilibrium (with a small variability of not more than +/- 243 mmHg/year)
        Ferdi, is a “seesaw” on a playground also more or less in equilibrium (with a small variability of not more than +/- 3 feet/day)
        Your silliness is irritating, Ferdinand.

        …… adding 4 ppmv extra CO2 per year to a system ………..does increase the CO2 level in the atmosphere with 2 ppmv/year,

        Ferdinand, you are again trying to bedazzle my mind with your mimicry of the results of “reverse” mathematical calculations ….. that probably originated with James Hansen et el. Taking the average 2 ppm yearly increase from the Mauna Loa record and back calculating it to get an average 20 billion tons of CO2 emissions by humans just to blame them for CAGW by fictitiously associating it with the “fuzzy” math calculated average global temperature increases is “junk science” and all part of the Great Global Warming Scam..

        No matter if that total natural part is 100 ppmv in and 102 ppmv out ….. or 1000 ppmv in and 1002 ppmv out over a year: the net contribution of nature to the increase is null, zero, nada…

        Ferdinand, framing your argument with selected values does nothing to DISPROVE the fact that the context of your commentary is little more than concocted tripe n’ piffle which has no basis in factual science.

      • Samuel,
        Missed this one…
        That humans are the cause of the increase in the atmosphere, despite the natural variability is as clear as that there is a sea level increase, despite the huge variability caused by waves, tides and storms. The difference is that one only needs 3 years to see that the CO2 increase is larger than the natural noise and that you need over 25 years to separate the small increase in sea level from the huge natural noise.
        Human emissions are not based on Mauna Loa data, they are based on fossil fuel sales (taxes!), with a higher probability of underestimates than overestimates, due to under the counter sales… Increase in the atmosphere is based on ice cores before 1959, Mauna Loa and other stations thereafter. Over the full period since at least 1900, the increase in the atmosphere is about halve the emissions.
        The air pressure does pump air in and out the upper snow/firn layer, but as far as I know, that doesn’t change the CO2 levels in the snow/firn. Measurements at several places show practically the same CO2 level in firn below the surface: 2 ppmv less at 5 m depth for Law Dome.

      • Thanks, Polycritic. First, Dr. Jaworowski seems totally unaware that there is a difference between air age and ice age.
        Next, Dr. J says that this the following is a “lie” …

        1. Carbon dioxide, the most important anthropogenic greenhouse gas, increased markedly as a result of human activities, and its atmospheric concentration of 379 ppmv (parts per million, by volume) in 2005 by far exceeded the natural range of 180 to 300 ppmv over the last 650,000 years.

        Next, Dr. J misrepresents the effect of additional inputs to a system at equilibrium, making the ludicrous argument that:

        … 97% of the total annual emission of CO2 into the atmosphere comes from natural emissions of the land and sea; human beings add a mere 3%. This man-made 3% of CO2 emissions is responsible for a tiny fraction of the total greenhouse effect, probably close to 0.12%. Propositions of changing, or rather destroying, the global energy system because of this tiny human contribution, in face of the large short-term and long-term natural fluctuations of atmospheric CO2, are utterly irresponsible.

        If you can’t see the huge fallacy in that argument, you need to contemplate the meaning of “equilibrium”. Here’s an example. If I make $100,000 per year and spend $100K per year, my net worth doesn’t change. My bank balance is at equilibrium.
        But if my expenses stay the same and I add “a mere 3%” to my income, my net worth will be 150% of the equilibrium value within a mere 17 years … and since Dr. J doesn’t seem to understand this, I fear my respect for him couldn’t go much lower.
        Finally, look again at my graph above. Consider the agreement between the different ice core records from different locations done at different times by different investigators. Then consider the changes in CO2 from earlier periods at the terminations of the ice ages, again from different locations done at different times by different investigators … if you believe Jaworowski, the only tenable conclusion is that these are all “lies” put together by some wide-spread collection of conspiratorial scientists.
        You’re free to believe Dr. Jaworowski if you wish. To me, I fear that he’s a crackpot. There are a whole host of valid arguments against the catastrophic anthropogenic global warming hypothesis. Dr. J’s arguments are not among them.
        w.

      • Willis
        You say

        You’re free to believe Dr. Jaworowski if you wish. To me, I fear that he’s a crackpot.

        I take severe offence at that especially when it comes from a nonentity!
        I consider it a great honour to have been associated with the late Zbigniew ‘Zeb’ Jaworowsk throughout the final decades of his life. He is the ‘father’ of ice core studies who conducted dozens of field trips to obtain ice cores, and he devised most of the methods now used to analyse ice cores. Hence, his anger when he learned of – what he considered to be – the abuse of the methods he devised for paleo-climate studies.
        Zeb provides the best summary of problems with ice core data in his Statement written for the Hearing before the US Senate Committee on Commerce, Science, and Transportation. It is titled “Climate Change: Incorrect information on pre-industrial CO2” and is dated March 19, 2004. As several have already said, it can be read at
        http://www.warwickhughes.com/icecore/
        Zeb’s severely failing health meant he could only submit the statement in written form.
        Similarly, his failing health prevented his attending the Heartland 1 Conference and he asked me to present his paper there so I did.
        I first became associated with him when the UN appointed him to determine the world-wide dispersal of radionucleatides from the Chernobyl nuclear disaster. He was a professor from a communist country and the disaster was in a communist country at the height of the Cold War. But nobody questioned his appointment because he was unarguably the outstanding authority in his field. However, he feared that whatever he discovered could be portrayed as having partisan bias so he desired association for review of his work by someone in a coal industry (coal was a competitor to nuclear) from the Western side of the Iron Curtain. I was a material scientist at the UK’s Coal Research Establishment (but not then the Senior Material Scientist) and filled that role. He was a communist and atheist while I am a socialist and Accredited Methodist Preacher, so we could be thought to be natural ‘enemies’, but I consider it a great honour that we were friends.
        And that friendship is why I add this background information in hope that I honour the name of the great Zbigniew Jaworowski. A great man, a true scientist and a sorely missed friend, who is much maligned by crackpots.
        Richard

      • Richard,
        As far as I have read, the late Jaworowski was a specialist in metal ions (radioactive or not) in ice cores, which is important for following the migration of the fallout of Tsjernobyl.
        I don’t know if he ever looked at CO2 in his own ice core measurements (if he did any), but it seems to me that many of his objections were written before 1992 with his knowledge of that time.
        Where he goes completely wrong is that he repeats all his earlier objections in 2004, while a lot of work was done on ice cores, especially the work of Etheridge e.a. on three Law Dome ice cores, published in 1996.
        Etheridge explicitly answered a lot of objections posed by Jaworowski in 1992 (without mentioning his name): he drilled three cores with three different methods (wet and dry) – no difference. He measured CO2 in firn top down until full ice and both firn and ice at the closing transect. He could calculate the theoretical migration in firn, the distribution of gas age in firn and ice and the gas age – ice age difference and he could compare that to what was really measured in the cores. That includes a direct overlap of ~20 years between CO2 levels in the ice and in the atmosphere.
        Thus sorry, the late Jaworowski may have been an expert on ice cores up to 1992, but all what he wrote after that was completely outdated.
        Some of what he wrote was already wrong at the moment that he wrote it, like the gas age – ice age difference which doesn’t exist according to him, or the migration of CO2 from low levels to high levels. Anyone with the slightest understanding of diffusion knows that this can’t be right.
        These last items closed the door for me: his knowledge in the past might have been superb, but what he wrote after 1992 is garbage…

        • Thus sorry, the late Jaworowski may have been an expert on ice cores up to 1992, but all what he wrote after that was completely outdated.
          Some of what he wrote was already wrong at the moment that he wrote it, like the gas age – ice age difference which doesn’t exist according to him, or the migration of CO2 from low levels to high levels. Anyone with the slightest understanding of diffusion knows that this can’t be right.
          These last items closed the door for me: his knowledge in the past might have been superb, but what he wrote after 1992 is garbage…

          Seems like he is the perfect and most excellent argument AGAINST the hollowed (er, hallowed) ideals of academic tenure and indentured (er, endowed) chairs of permanently enshrined professors at outhouses (er, our houses) of academic domains.

      • There is no “human signature” that is visible or measurable in the official record (Mauna Loa) of increases and/or decreases in atmospheric CO2 ppm quantities. Assuming that there is such a “signature” is junk science. ….. (And “NO”, ??x??, …. there is no H-Pyron signature.)
        And humanity being responsible for a mere 3% of the total yearly emissions of CO2 into the atmosphere ….. is akin to …. sibling family members being responsible for a mere 3% of the total yearly deposits of money into the parental “expense-paying” Checking Account at a Bank.
        And neither of the aforesaid three (3) percent’ers (3% ) are cumulative from one year to the next in their respective depository. Mother Nature does not pay “interest” on deposits.

      • richardscourtney January 26, 2015 at 1:56 am

        Willis
        You say

        You’re free to believe Dr. Jaworowski if you wish. To me, I fear that he’s a crackpot.

        I take severe offence at that especially when it comes from a nonentity!
        I consider it a great honour to have been associated with the late Zbigniew ‘Zeb’ Jaworowsk throughout the final decades of his life. He is the ‘father’ of ice core studies who conducted dozens of field trips to obtain ice cores, and he devised most of the methods now used to analyse ice cores. Hence, his anger when he learned of – what he considered to be – the abuse of the methods he devised for paleo-climate studies.

        Richard, when you get done with taking “severe offense”, perhaps you could point us to some of his papers outlining his devising of “most of the methods now used to analyze ice cores”. I looked hard, and I could find nothing of note. I did find a couple of papers about heavy metals or radioactivity in ice cores, but nothing about how he devised the methods in use.
        Next, both Ferdinand and I have pointed out specific problems with his claims. Rather than refute a single one of them, you go on and on about your friendship with him, and how you presented his paper, and how he was an atheist and you weren’t, and how “the UN appointed him to determine the world-wide dispersal of radionucleatides from the Chernobyl nuclear disaster” … so freakin’ what? How about you talk about the science and not the man?
        Here’s an example. He says that the following statement is totally erroneous:

        The age of gases in the air bubbles is much younger than the age of the ice in which they are entrapped (Oeschger et al. 1985), the age difference ranging from several tens to several ten-thousands of years.

        But our guest author has just explained in great detail that not only is that statement true, he’s laid out the physical reasons why it is true. So perhaps you can explain how Zeb is correct and the rest of the scientific world, including our guest author and most of the commenters, is wrong about their being a difference between ice age and air age.
        If you want to get some traction, deal with the issues that Ferdinand and I identified. I don’t care if Jaworowski studied radiation or whether he provided blankets and dinner for the homeless or whether he is an atheist or a Buddhist monk, that has nothing to do with the errors in his claims.
        Look, richard, I’m sorry I called him a crackpot, that was a bit over the top. Let me adopt the lovely Nixonian language and say that that statement is “no longer operative”.
        But as Ferdinand has pointed out, even when presented with evidence late in his life that some of his views were wrong, he refused to face the facts … I’ll leave you to decide whether that is science or not.
        w.

      • @ Ferdinand Engelbeen:January 26, 2015 at 7:54 am

        Samuel, it seems quite difficult to understand that a small disturbance of a few % of what is a natural cycle may give a large increase (or decrease) over time

        Ferdinand, you are “talking trash” …. and you know it. Or are you really that miseducated?
        The emissions of CO2 into the atmosphere, …. from all sources, ….. has nothing whatsoever to do with “natural cycles” … EXCEPT FOR the “ natural seasonal cycles” associated with the earth’s axial rotation and orbital revolution. There are no other “natural cycles” involved, … NONE, NADA, ZERO, ZILCH.

        Something many housewives have no problem with: if they slightly spend more money than their household budget allows, they know they are getting into trouble…

        Ferdinand, your childish analogy is an insult to even the Learning Disabled.
        “DUH”, you have never been married, have you? Or at least, never been “head-of-household” revenues and expenses.
        And ps, housewives never get into trouble as long as they are still capable and/or willing to exhale copious amounts of CO2 into the atmosphere whenever sexually excited.

      • Samuel,
        Everybody on this blog knows that I have lot of patience and that I try in many ways to show people what is true and what is false in someone’s reasoning. But even my patience has its limits. If we are discussing CO2 levels in the atmosphere the 3% rule is popping up again and again. That is NOT an argument at all, it is as stupid as my example (and yes I am over 40 years married and still happy, as we both like to spend as much as we earn).
        If you don’t (want to) understand that adding 4 ppmv extra CO2 per year to a system which is more or less in equilibrium (with a small variability of not more than +/- 1 ppmv/year) does increase the CO2 level in the atmosphere with 2 ppmv/year, then no argument of mine can convince you that the increase can’t be natural, whatever the individual sources (volcanoes, burning forests, heating oceans) or sinks (polar oceans, greening of the planet) do over a year. No matter if that total natural part is 100 ppmv in and 102 ppmv out or 1000 ppmv in and 1002 ppmv out over a year: the net contribution of nature to the increase is null, zero, nada…

    • If this graph is accurate, where are the variations of CO2 levels for the medieval warm period and the little ice age? No variation of CO2 with wide variation of climate???

      • One need high resolution ice cores to see the change, which is visible in de Law Dome DSS core (~20 year resolution over the past 1,000 years):
        http://www.ferdinand-engelbeen.be/klimaat/klim_img/law_dome_1000yr.jpg
        The drop of ~6 ppmv lags the drop in temperature (based on the Moberg reconstruction) of ~0.8°C with about 50 years. The ratio again is the same as for the glacial – interglacial changes: about 8 ppmv/°C.
        That also means that the warming after the LIA is only good for maximum 8 ppmv CO2 increase, the rest of the 110 ppmv is from human emissions…

      • Engineer Ron: the Vostok and Dome C ice cores show a surprising linear relationship between temperature and CO2 of 8 ppm/K. The transient responses over the seasons is globally ~5 ppmv/K and the 2-3 year fluctuations (Pinatubo, El Niño) also are 4-5 ppmv/K.
        Thus while many underlying reactions may be far from linear, the overall T-CO2 relationship is quite linear in periods of a few decades to many thousands of years, be it with variable lag times for CO2 after T over spans of 12 K and changes of 100 ppmv in the past.
        All ice core resolutions are more than sufficient to detect a peak of 110 ppmv like the current one if that occurred anywhere in the past 800 kyears. There was no such peak and the current increase is far beyond the equilibrium between oceans and atmosphere (by Henry’s law) for the current temperature.

      • Samuel, if the MWP was as warm or warmer than the current period, the current warming was not more than the cooling between MWP and LIA. Thus the same increase in CO2 of 8 ppmv as the drop in CO2 between MWP and LIA…
        Except if you prefer Mann’s HS shape: then there was hardly a drop in temperature between MWP and LIA and a huge increase in recent temperature in recent times…

      • @ Ferdinand Engelbeen:January 26, 2015 2015 at 8:12 am

        Samuel, if the MWP was as warm or warmer than the current period, the current warming was not more than the cooling between MWP and LIA.

        Ferdinand, now that was a truly remarkable deduction. Did you thunk it up all by yourself, … or did you have help with it?

      • Samuel, if the changes in temperature were equal, why would a cooling of app. 0.8°C give a drop of 6 ppmv CO2 (with a lag of ~50 years), but a warming of 0.8°C would give an increase of (unfinished) 50 ppmv CO2 in the same ice core? Please elaborate…

  28. Recall Gibbs-Duhem equation and the transformations of sediment (snow, very low density) into a sedimentary rock (firn, middle-range densities) and onto a metamorphic rock (glacial ice, high density, though less than carbonate and silicate rocks) and the retrograde metamorphism of glacial ice back to firn (retrograde meta-sedimentary rock) then re-metamorphism back to glacial ice (multi cycle metamorphic rock) with the principle of isostacy in cycles over geologic time.

  29. Well, as expected, here my comment…
    1. The resolution
    There is an essential error in Ronald Voisin’s reasoning: at the start of bubble closing at sufficient depth (~70 m and deeper, depending of temperature), the average age of the air in the pores is not 20 years (for the high accumulation Law Dome cores), but only 7 years younger than the air at the surface. That is because there is a lot of time for migration during 40 years, which only is reduced when the pores are getting smaller and smaller with increased density of the ice with depth.
    Then we have the time needed for fully closing all the bubbles, which is for the Law Dome cores between 72 and 80 meters, adding again a few years to the average, but also with slightly different age in different bubbles. That makes that the average gas age at full closing depth is 10 years younger than the age of the ice itself at the same depth and the gas composition is a mixture of ~10 years of air:
    http://www.ferdinand-engelbeen.be/klimaat/klim_img/law_dome_overlap.jpg
    Thus the resolution of the ice core gas composition is much better than expected from the time needed to start bubble closing.
    Much depends of the amount of snow accumulation: at closing depth, the gas composition is near the same for all ice cores, no matter the accumulation, but the ice age fully depends of the accumulation rate: 40 years for the 1.2 m ice equivalent/year accumulation at Law Dome to several hundreds of years for Vostok or Dome C where only a few mm/year snow is falling.
    The accumulation rate also influences the time needed to fully closing all bubbles. For the high accumulation Law Dome ice cores that leads to a resolution of ~10 years. For Dome C that is ~560 years and for Vostok that is ~600 years.
    The drawback of high accumulation is that the number of years of the ice core before reaching rock bottom is shorter: 150 years for Law Dome, 1,000 years for the Law Dome DSS core (taken downslope, ~20 years resolution), 70,000 years for Taylor Dome (~40 years resolution) and 420,000 years, resp. 800,000 years for Vostok and Dome C.
    2. The peak detection
    The accuracy of ice core CO2 measurements for multiple samples taken at the same part of an ice core by different laboratories is 1.2 ppmv (1 sigma). Samples taken from different ice cores with the same (calculated) gas age differ not more than 5 ppmv from each other.
    Any cyclic peak which occurs with a frequency smaller than the resolution will not be detected, as the positives and negatives compensate each other. But any one-way peak which is 2 ppmv or more sustained over the full period of resolution or a one-year peak of twice the resolution in years will be detected.
    The current increase is ~110 ppmv over 160 years. Even within the worst resolution ice core (Vostok) a one-way peak of 110 ppmv would be detected as a peak of over 25 ppmv, which is a quarter of the measured glacial-interglacial changes in CO2. Even if the current increase was a part of a natural cycle of over 600 years (there is currently no sign of increase speed reduction), that would be detected in all ice cores…
    3. Ice core CO2 migration
    There is no direct measurable CO2 migration in ice cores.
    One has made several attempts to calculate a theoretical migration: one of the last was by looking at the (coastal) Siple Dome ice core which shows several re-melt layers where nearby CO2 is enhanced. The theoretical migration of this relative “warm” ice core (-22°C) is a broadening of the resolution from 20 to 22 years at middle depth and from 20 to 40 years at full depth (~70 kyear). No big deal. For the cold (-40°C) inland ice cores there is no measurable migration at all over 420 and 800 kyear.
    Important is to note that migration of CO2 in ice cores does influence the resolution, but that doesn’t change the average over the resolution period.
    4. The stomata data show more peaks
    Stomata (index) data are proxies: there are less stomata when the average CO2 level in the previous growing season was higher. The main problem, besides confounding factors, is that these are local CO2 levels over land, not “background” CO2 levels in the atmosphere as is the case for Antarctic ice cores. That means that one can calibrate the stomata index data over the past century against ice core CO2 and direct measurements, but nobody knows how the local CO2 levels changed with land use changes in the main wind direction, including changes of the main wind direction itself over certain periods (MWP-LIA).
    Thus while stomata index data have a better resolution than ice cores, the variability is mainly local variability and if the average of the stomata index data differs from the ice cores over the ice core resolution, then the stomata data are certainly biased…
    5. The objections of the late Jaworowski
    I have looked at the objections of the late Jaworowski. I regret to say that what he wrote is completely outdated since 1996, when Etheridge e.a. published their work on three ice cores at Law Dome. Moreover, some of what he wrote is physically impossible (migration of CO2 from inside the ice core at 180 ppmv to the outside at 380 ppmv)…
    See: http://www.ferdinand-engelbeen.be/klimaat/jaworowski.html
    More about ice core measurements:
    http://courses.washington.edu/proxies/GHG.pdf

      • That will make most difference at full depth, where the resolution doubled, but hardly any at middle depth. Even so a 10 ppmv peak sustained over 10 years still would be detected as a 2 ppmv peak in the Siple ice core…

      • @ Ferdinand Engelbeen:January 26, 2015 2015 2015 at 8:30 am

        Samuel, the last peak is the one that differs from the others:

        Ferdinand, getta clue, ….. all the peaks are wrong, ….. the above graph I citted is FUBAR.

    • Ferdi,
      As I said to Willis: Willis, the good fidelity you speak of can only exist while the atmospheric CO2 is higher than what has been previously recorded. As soon as it reverses, fidelity will be hugely compromised.

      • Ron, there were 8 large warming and cooling phases in the past 800 kyears ice core record. They all show the same ratio and trends of CO2 following the temperature changes. No difference in ratio, only some differences in lag time (far more during cooling than during warming). Seems to me that the ice core record is quite reliable.

      • Engineer Ron, there is no attenuation at all: the same ratio over 8 interglacials, with durations of 5000-10000 years. The lags are about 600 years during a warming period and several thousands of years into a cooler period. There is no measurable lag in the CH4 record. If migration was a problem, CH4 would be more problematic than CO2.

      • Samuel, the last peak is the one that differs from the others: 110 ppmv above the historical equilibrium for the current temperature. Humans have added about 200 ppmv to the atmosphere. You may be sure that the extra 110 ppmv is not caused by humans, but I haven’t seen any good argument for that: all alternatives I have heard of violate one or more observations…
        In short, that is a lost battle and a very bad argument in the climate wars…
        The main battle is about the sensitivity of temperature for a CO2 doubling…

    • Ferdinand,
      I am not so quick to dismiss Jaworowski who has decades of ice core experience and expertise internationally recognized and who has been poor mouthed by those who poor mouth anyone who does not conform to the cult codes.
      In fact, the odds are that any who are vilified by the cult must be correct.

      • mpainter, the late Jaworowski may have had a lot of experience, but that ended around 1992, where the work of Etheridge e.a. of 1996 shows that he was wrong on several points.
        I lost any confidence in him when he said that CO2 migrates from 180 ppmv within an ice core towards 380 ppmv of outside air and when he “showed” that Neftel “cheated” by an “arbitrary” move of 83 years to align the CO2 in the ice gas bubbles with the Mauna Loa data: he simply looked at the wrong column in Neftel’s table of ice age and gas age.
        When I mailed him about that point, he answered that there was no difference between gas age and ice age as all ice cores have re-melt layers which isolate the pores from the atmosphere. Which is simply not true.

      • mpainter January 25, 2015 at 2:29 pm

        Ferdinand,
        I am not so quick to dismiss Jaworowski who has decades of ice core experience and expertise internationally recognized …

        Jaworowski was an expert, but not about CO2. He was a physicist whose area of expertise was radiation and radiation risks. As far as I’ve been able to find out, he never did any original research regarding ice cores. Instead, he wrote a few papers that made a number of incorrect claims about errors in ice core analysis. I have pointed out some of his errors. Ferdinand has pointed out a number of others, in much greater detail and with a number of citations.
        If you disagree with either of us, telling us that “the odds are that any who are vilified by the cult must be correct” is a joke. And telling us (incorrectly) that he has “expertise internationally recognized” does nothing. That’s just an appeal to authority.
        If you think we’re wrong, then tell us exactly where you think Ferdinand or I are wrong … and don’t forget to quote the exact words you disagree with so we can understand your objections.
        w.

    • There is much discussion about local variability of CO2 invalidating plant stoma in these pages. NASA has just put up a satellite to look at local variability of CO2 so as to determine the sources of CO2 in the world. As I remember it, the entire scale of the global plot was a variation of about 10 ppmv! If the entire local variation can be put on a delta 10 ppmv scale how does such a small variability in any way invalidate plant stoma proxies?

      • I don’t know the volume of air measured above ground by the OCO-2 satellite (the Japanese satellite measures the whole air column), but the largest CO2 levels over ground are near the surface (1000 ppmv and more) and may change from 550 (and more) ppmv over the first 100 meters (and higher) at night under inversion and 280 ppmv during the day and full photosynthesis.
        The NASA satellite only measures around midday, which may give some impression over a year (or years), but that is not a local average.
        Local CO2 levels, even averaged over months show large swings (and extremes) and a ~40 ppmv local bias compared to “background” CO2 levels in 95% of the atmosphere. Here for Linden/Giessen (SW Germany), semi-rural, a few km from the original historical site where a long series in the 1940’s was taken:
        http://www.ferdinand-engelbeen.be/klimaat/klim_img/giessen_mlo_monthly.jpg

    • Why do you think a bubble closing stops diffusion? Put air vs helium in a balloon and wait for a couple of days, to see how gases diffuse through solids.

      • Tony

        Why do you think a bubble closing stops diffusion? Put air vs helium in a balloon and wait for a couple of days, to see how gases diffuse through solids.

        There are valid arguments and discussions about closing ice bubbles slowly over time due to constantly changing gas levels and temperatures over 10-40 years.
        But please, don’t confound those differences with a 2He4 single gas molecule under relatively high differential pressure diffusing through a single closed rubber envelope (compared to a “air” molecule of 80-20 (14N28)2 (molecular weight = 28) and 2(16O32) molecules)!
        Even if your comparison were valid, it goes the other way: CO2 (with a single tightly bound molecule of 12 + 16 + 16) is only slightly harder at getting through a crack in the top of an air bubble in ice with a double “air” molecule of either N2 or O2 or Ar, isn’t it?

      • RACookPE1978, My point is that if gases can diffuse through the solid walls of a balloon over a couple of days, there will be significant diffusion through solid ice bubble walls over hundreds of years. My point in mentioning helium is simply that rates of diffusion are different and the deflation of a balloon is not simply caused by gases leaking.
        Diffusion through solid bubble walls resulting from differences in concentrations of CO2 and other gases will result in reducing differences in gas concentration differences at different levels. That is, CO2 concentrations will appear more uniform over long periods. Ice bubbles do not “lock in” atmospheric gas concentrations.

      • Tony,
        There is no migration of CO2 through the ice matrix. There may be some migration through the liquid-like layer between ice crystals (none below -32°C) and through channels caused by impurities, but even that is not measurable.
        Theoretical migration in a relative “warm” (-23°C) ice core was a matter of a few centimeters after 70,000 years.
        If there was any substantial migration in the much colder (-40 °C) high altitude inland ice cores, then the ratio between CO2 changes and temperature changes (~8 ppmv/°C) would fade with each interglacial back in time each 100,000 years apart. But that is not observed.
        BTW rubber and most plastics are not that solid, a Coke PET bottle loses its CO2 after many months. Metals (and ice) have a much lower permeability but even then the smallest molecules like helium and hydrogen may migrate into metals and hydrogen at elevated temperature may remove the carbon in steel (as methane), reducing its strength at high temperatures…

        • Ferdinand

          BTW rubber and most plastics are not that solid, a Coke PET bottle loses its CO2 after many months. Metals (and ice) have a much lower permeability but even then the smallest molecules like helium and hydrogen may migrate into metals and hydrogen at elevated temperature may remove the carbon in steel (as methane)

          Well, hydrogen as a gas at room temperature does leak directly through steel pipe walls rather quickly (and dangerously) at room temperature and relatively low pressures. Liquid hydrogen not so much, but then the pipe walls are also very, cold and so the pipe wall interstitial distances are closer together. (Liquid hydrogen is also at lower pressures, which helps reduce through-wall leakage. Methane? Never so much leakage as to be critical – though there is some. (Methane is CH4 – so it is only 16 molecular weight. ) Hydrogen embrittlement is a serious problem that must be addressed in every pipe weld by metal selection, weld preheat and post heat treatments, rigorous weld rod coatings and weld rod temperature and humidity controls, and weld bare metal cleanliness and grinding controls.

      • policycritic, as far as I can see, Jaworowski never mentioned or responded to the findings of Etheridge and others about ice cores in 1996 and later. His 1997 and 2004 writings were near an exact copy of his earlier writings of 1992. Seems quite strange for someone who is an expert in ice cores…
        It seems quite difficult to find an independent source of what he has really done on ice cores, his biography at Wiki is one big flame war… But his own c.v. doesn’t mention anything about CO2 in ice cores:
        http://www.bibliotecapleyades.net/ciencia/ciencia_sol29.htm

    • @Ferdinand Engelbeen and Willis Eschenbach,
      I’m curious. Did Dr. Jaworowski respond to Etheridge’s 1996 findings? Surely, he must have been aware of them since Jaworowski, per his Senate statement, was working on ice cores, or what he calls “glacier findings,” with the “Norwegian Polar Research Institute in Oslo, and in the Japanese National Institute of Polar Research in Tokyo,” and he writes:

      In this period I studied the effects of climatic change on polar regions, and the reliability of glacier studies for estimation of CO2 concentration in the ancient atmosphere.

      Thanks in advance for any enlightenment.

  30. Prof Murry Salby’s your man for this.
    One observation is than the instant an icecore is drawn to the surface it outgasses most of the trapped air.
    So whereas the O16/O18 in the ice itself is a reasonable proxy for temperature, the measurement of the CO2 in the residual air pockets is clearly going to be very low. In fact one might regard it as a first differential of [CO2] with Time.
    Thus the signature of CO2 will tell when it rose or fell but little about actual amounts in the air at the time.
    They would have been much much higher. The deeper the icecore the greater amount of trapped air is lost on extraction. We are possibly looking at 10, 20,100x even 1000x more CO2 in the air than that calculated from the air bubbles.

    • Sorry Philip, but Prof. Salby is completely lost on this point. What he says is simply impossible. As said before: migration does lower the differences, but it doesn’t change the average over the period of resolution.
      If Salby was right and the peak of the last interglacial of 100 kyear ago was 10 times higher than measured, then the 3000 ppmv (according to Salby) was spread over the rest of the 100,000 years before and after (and that of the second interglacial before too was migrated). That means that the measured 180 ppmv in glacial times was negative, effectively killing near all life on earth (even below 180 ppmv a lot of plants have trouble surviving).
      Secondly, migration only stops if all differences are gone. There are 8 similar peaks of around 300 ppmv CO2 over interglacials, each 100,000 years further back in time. If there was huge migration, then the first peak was underestimated with a factor 10, the second a factor 100, and the last of 800 kyear ago was underestimated a factor 10^8. That is more than all directly available carbon on earth, including oceans and carbonate rocks…

      • Ferdi, Someone is lost for sure. Your negative concentration arises only when you think all the CO2 is trapped even in spatially in the wrong place. Philip Foster is correct – the CO2 is lost, not just because of the initial attenuation but due to outgassing when removed from a high-pressure environment to atmospheric pressure.

      • Ron, Prof. Salby was talking about migration of CO2 in the ice core itself, not about loss of CO2 by outgassing…
        Theoretically there may be a loss of pressure through cracks or so during relaxation of the ice core after drilling. But I don’t see a preferential escape of CO2 out of the high pressure air bubble in the ice core, to the contrary: CO2 has more affection to water molecules than O2 or N2. The same for decomposing clathrates: N2 and O2 clathrates decompose already at much lower temperatures than CO2 clathrates. Thus if there is loss of air through cracks, it is O2/N2 first, CO2 second, leading to too high CO2 levels, not too low.
        Further, after pressure equilibration, if there are cracks in the ice, the migration would be from the outside near 400 ppmv to the inside 180-300 ppmv, not reverse…

      • “Thus if there is loss of air through cracks, it is O2/N2 first, CO2 second…”
        Does not follow. Depends on relative concentration.

      • Bart, if the loss is pure mechanical from the higher inside bubble pressure to the atmosphere, then there is no change in composition. If the loss is slow enough, there is a higher possibility of CO2 remaining in the bubble and/or the cracks as it has a much higher affinity to water or water-like layer at the ice surface than O2 or N2. At measuring time, under vacuum (which removes practically all liquid water), that leads to relative higher CO2 levels, not lower.
        Once the internal and external pressures are in equilibrium, the migration is from highest to lowest levels, but as the ice cores all show 180-300 ppmv in the past, and recent to current atmosphere is 380-400 ppmv, any migration is from the outside to the inside, not reverse…

      • Ron, I have been listing several times to his speech in Germany (not easy to follow by times) and was in London to listen to him in person. He really thinks that there is high migration of CO2 in ice cores and therefore the real CO2 levels at the previous interglacial of 100,000 years ago were 10 times (London, 15 times in his German speech) underestimated by the current measurements. Which is physically impossible…

  31. No one seems to be addressing the affects of organic processes on the CO2 distribution. Organic processes could tend to pump CO2 towards the surface, just as nutrients tend to be pumped towards the surface in soils. Of course, research into this is unlikely to be carried out in an unbiased manor considering the political beliefs of those capable of undertaking such research.

  32. The “recordation” would be a mathematical convolution of two functions.
    1. The input shape, that is the shape of figure 1. The removal shape would likely be logarithmic decay, strongly influenced by the response times of the sinks (that are always operating.
    2. The firn closure diffusion gradient (describe by WE above). Mostly vertical, but near the ice-firn interface becoming non-linear as the bubbles closed and the diffusion coefficient approaches near zero.

  33. “However, do WUWT readers see any problems with this “ideal” characterization of the ice-core recordation process?”
    Yes, and a big one, though it appears little appreciated.
    If, as the ice cores suggest, CO2 was maintained in a very narrow band for thousands of years, then there had to be a relatively high bandwidth regulatory system keeping it there. Low bandwidth systems do not maintain tight regulation. In a timeline of less than the dominant time constant, they wander, like the particles in a fluid undergoing Brownian motion, with RMS variability evolving as the square root ot time.
    A high bandwidth regulatory system would shrug off human inputs as just a tiny perturbation. It is inconsistent to claim high bandwidth and low bandwidth simultaneously. Either the ice core estimates are wrong, or human inputs cannot have a significant effect.

    • Bart, the “high” bandwith needed to show the current human contribution is in the order of 50 years. Slow enough to show natural variations of seasonal to 2-3 years caused by temperature variations and more than fast enough to show a quite linear response of CO2 to temperature of about 8 ppmv/K over periods of 5,000 years (glacial-interglacial transition)…

      • To clarify for others, Ferdinand is not even addressing the question. This is an issue of regulation, not of observation.

      • Bart, let us say that your interpretation of the regulation doesn’t match the observations (not one of the many observations to be clear). Thus I suppose that your interpretation of the regulation is wrong…

      • Bart are you saying that an e-fold decay rate of 50 years (as currently observed in the atmosphere) for excess CO2 above equilibrium is not fast enough to tie the historical CO2 levels to temperature changes of 0.0024°C/year?

      • I am saying it is not possible for the Earth to be able to regulate CO2 levels to within such a small range over millennia, then be unable to cope with the meager inflows from human activity. It is inconsistent for such a high bandwidth regulatory system to be so sensitive to such a minor disturbance.

      • Bart, that is a matter of time frame: with the observed 50 year e-fold rate it is perfectly possible to let CO2 levels track temperature changes of ~100 ppmv/12°C within a very narrow band over a time frame of 5,000 years. Much faster changes of temperature like seasonal or 2-3 years would give disturbances of the CO2 level in the order of 4-5 ppmv/°C which will go unnoticed in any ice core as the resolution of at least 10 years averages them out. Sustained drops of temperature like the MWP-LIA transition are noticed in the relative high (20 years) resolution downslope ice core of Law Dome as a drop of ~6 ppmv.
        The current human emissions exceed the capability of the sinks to remove the excess CO2 in the same year as the emissions. That is what is observed. But the 110 ppmv increase over the past 160 years is by far large and long enough to be noticed in every ice core, even with the worst resolution.

      • You’re still not getting it, Ferdinand. There are random fluctuations in CO2 input, from droughts to unchecked biological growth, from volcanoes to wildfires, to asteroid strikes to deep sea burps… and more. All of these vagaries feed into the overall level.
        The random nature of these inputs would send the atmospheric level off in a widely ranging random walk if the system were not highly regulated with a wideband regulation process. Such tight regulation would also suppress any minor human inputs from significantly affecting things.
        You can’t have both – high sensitivity to human inputs and low sensitivity to random, natural inputs. The sensitivity is either high or low for both. A millennium of tightly controlled variation is inconsistent with significant human forcing.

      • Bart, you still are not getting it: as long as the random fluctuations are within reasonable borders, they are returning to the temperature dictated equilibrium within the resolution of the measurements.
        Most seasonal to short term variations are within that time frame:
        – Seasonal changes are about 5 ppmv/°C within months in extra-tropical vegetation in the NH but level off over a year.
        – Short term changes caused by El Niño’s, including drought periods in the tropic vegetation are 4-5 ppmv/°C but level off over periods of 2-3 years.
        – Current continuous degassing of the equatorial upwelling places are more than compensated by continuous sinks near the poles.
        – Volcanic emissions in general are continuous, small (~0.1 GtC/year) and fully absorbed in the current time frame.
        What is not in that time frame:
        – Huge volcanic emissions over long time frames, like the Deccan traps which did spew lots of CO2 over 10,000’s of years.
        – Huge meteor impacts (like Yukatan)
        – Longer term changes in deep ocean upwelling
        – Human emissions
        The current short term (seasonal to 2-3 years) are measureable in direct observations, but not in ice cores.
        Long term changes caused by changes in deep ocean upwelling and human emissions are observable both in direct measurements and in ice cores. The former as a continuous following of CO2 after T over millennia, the latter as an increase of CO2 beyond the long term temperature dictated equilibrium between oceans and atmosphere.
        The 50 years e-fold decay rate for any disturbance around the long term (deep ocean) temperature equilibrium is not fast enough to level off short term reactions of vegetation on temperature and drought or human emissions. But it is by far fast enough to get CO2 following temperature on glacial – interglacial scales.
        The short term reactions on temperature are mainly from vegetation, the reaction of the deep oceans on temperature is much slower. Different processes with different reaction times (filtering) at work.

      • You don’t get it. Keep watching. Human emissions keep accelerating. CO2 maintains a steady space. Temperatures lead CO2 by 90 deg. There is no doubt about it. This rise is almost entirely natural.

    • Sorry, mostly from carbonates + highly acidic Icelandic volcanic dust, both much less or absent in Antarctic ice cores… Reason why Greenland ice cores are not used for CO2 measurements.

    • Only in “warm” coastal ice cores: a broadening of the resolution from 20 to 22 years at middle depth and from 20 to 40 years at full depth (60-70 kyears), not measurable in the much colder inland ice cores.

  34. Ronald Voisin – Thanks for your interesting article. I read up to where you ask “do WUWT readers see any problems”, and thought it best if I answered at that point rather than read on and have my mind corrupted (or corrected) by others’ thoughts.. I see two basic problems:
    1. When the firn is closing over a say ~40-year period, I believe the assumption is that the air still mixes freely with the atmosphere right up to the moment of closure. ie, that the firn is permeable for the duration. This then implies that the air captured in the ice is 40 years younger than the ice. The reality surely is that the firn is porous for the duration, but not necessarily permeable, ie, some air bubbles will by chance be closed off early. Thus when an ice core is analysed, the trapped air molecules at each depth will come from a period of up to ~40 years, not from a point in time.
    2. I believe there is an assumption that the air molecules trapped in the ice cannot migrate. ie, that the air molecules cannot move through ice. When the air-bubble-filled firn condenses into ice, the air bubbles are absorbed into the ice. In the absorption process, the air molecules must travel a distance x into the ice where x must be >0 therwise there would still be an air bubble. So the air molecules must be able to move through ice. That assumption therefore is necessarily incorrect.

    • 1. Is right, except that all pores remain open until closing depth, which is at 72 meter and below (depending of temperature). That means that the first bubbles close at 72 m (average 7 years older than atmosphere) the last at ~85 m (10 years later), which gives an average 10 years resolution and an ice-gas difference of ~40 years. The gas composition in the firn (and ice) was measured by Etheridge e.a. top down, layer by layer, until full ice. Both firn/pore gas and fully closed bubbles had the same composition at the same depth.
      2. Is not right: the bubbles are not absorbed in the ice, the pores simply are closed by ice crystals as they are reduced in size by the increasing ice pressure above them. The air bubbles remain in place and visible until the pressure is high enough to form clathrates, but that is a different story.

      • Thanks. One problem: You say “Both firn/pore gas and fully closed bubbles had the same composition at the same depth”. But they should have different compositions, because the fully closed bubbles would contain air from an earlier year than the firn/pore gas.

      • The measurement is not that fine that individual bubbles can be measured (in the period of drilling ~1995), so still a mixture of different gas bubble ages was measured with a difference of ~8 ppmv over 10 years time, rather small difference where the open pore gas age was measured at 1 meter intervals. Seems both probably too coarse to find a difference at that time…

      • We still have a problem. If they can’t separate them for meaurement then they can’t know that firn/pore gas and fully closed bubbles have the same composition.

      • Mike, to be more accurate: the average composition of the gas in the still open pores at 1 m intervals and the average composition of the air in the ice core bubbles at the same interval, measured by the same equipment (GC) was within the accuracy of the measurements (1.2 ppmv, 1 sigma) for the measurement accuracy of the 1990’s.
        I suppose that the current analytical equipment (total sublimation, mass spectrometer, only 8 grams of ice needed) can give a much finer resolution of the air composition in the ice, but taking representative small samples of the air in the open pores still would be a challenge.

    • Really? Most of that was already refuted by the work of Etheridge e.a. in 1996. Either Jaworowski hasn’t read any literature about ice cores after 1992, or he didn’t want to know what others have done in his former field of knowledge.
      Sorry, too many severe errors which are physically impossible by the late Jaworowski. Let him rest in peace, together with many of his ideas about ice cores…

      • Hello Ferdinand.
        You seem to have a good knowledge of ice cores and the ice core data.
        In that respect I am interested to learn and know one or two things, provided that you may help.
        I am interested about the ice core data regarding the last glacial period, the 100K long period.
        The two things I like to know are; how the age of ice is established for the sampling taken and what resolution of sampling used for CO2 data extraction, for the given period in question.
        Appreciated.
        cheers

  35. Until we have a billion dollar 100 year study to determine if CO2 in the ice is actually representative of the atmosphere at the time, since the only other proxies we have are so inaccurate as to be assumed intentionally contradictory, and whether the conversion from compacted snow to actual ice actually retains the proper ratio or if it simply squeezes out the CO2 increasing the immediate value in the lower levels of compacted snow resulting in a blurring and constant amplified level we’ll not actually get a damned bit of meaningful data out of ice cores to begin with. CO2 doesn’t freeze at all in the atmosphere and is one of the first gasses to respond to any input of energy in the conditions normally prevalent at glaciers because it remains in either a gaseous or soluble state, unlike water and even methane which can freeze within our atmosphere and ocean.
    So to even begin to calibrate what the actual level of CO2 in the atmosphere was at the time of snowfall we’d also need to know the temperature, wind speed, cosmic radiation level and amount of sunlight. Four datum we’re exceptionally unlikely to acquire – ever – which influence whether the CO2 stays with the water ice through the process into antiquity.
    And now we know that there are years where anywhere between one and fifteen years of data are lost due to melting – which would necessarily change the partial pressures of any trapped gasses below them in the ice.
    It’s a crap shoot to begin with – science for science’s sake. Still a better bet than people cooking the world due to cow farts…

  36. Not that you plan to cover any of my niggling little doubts regarding ice cores, Ronald, here is my short list.
    Your graphs list loose snow, firn and ice as the differing layers. Those layers are according to man’s touch, look and feel.
    Snow is deposited in layers over very large areas. Cores are not drilled over very large areas seeking confirmation or even correlation. Yet we are told that ice flows, downhill and away from pressure, giving us a concept that ice, firn and snow are plastic.
    The along come the ice core experts who point to a miniscule sample and insist they can identify atmospheric compositions from trapped bubbles.
    Solid pure ice is expanded slightly from liquid water and fills volumes based on H2O’s rigid molecular lattice structure hinged on hydrogen bonds with Oxygen. A structure leaving open conduits within pure solid ice with a rough gap of 2.76 angstroms.
    Add in impurities, e.g. say dust, sulfur, carbon, bird feces, polar bear debris, seal and Arctic fox debris causing ice to not organize properly.
    Carbon dioxide is roughly 3.3-3.4 angstroms in size.
    Until ice is pressed into a pure solid there are highways for CO2 to travel.
    Ice as coarse fallen snow is certainly not impervious to relatively easy CO2 migration. Just the action of barometric pressure changes will cause the firn and coarse ice to act like a bellows
    Ice as firn may look somewhat visibly solid to man, but at a molecular level easily allows CO2 migration. Migration not only in a vertical direction, but definitely along layers.
    At what pressure level does ice actually reach a solidity that is somewhat restricts CO2 movement? Whenever, ice core specialists point to ‘bubbles’ in their ice cores, I have doubts that the ice in question ever reached solidity.
    Which returns us to, just what has been performed, and replicated, to verify one ice core’s data with multiple ice cores over an area? Sample size, i.e. an area much larger than a football field,, cored in a consistent ordered sampling.

    • Have a look at:
      http://courses.washington.edu/proxies/GHG.pdf
      CO2 diffusion stops at an ice density of around 0.85 kg/dm3. That is between 72-90 m depth, depending of temperature or an ice pressure of 6-7 bar.
      In general multiple cores are taken at any site (Law Dome: 2 at the summit, 300 m of each other, one downslope at about 20 km from the others), but more important, many cores of quite different sites with enormous differences in accumulation rate, temperature, dust inclusions, etc. all show similar CO2 levels for the same average gas age within +/- 5 ppmv. Here for the past 10000 years:
      http://www.ferdinand-engelbeen.be/klimaat/klim_img/antarctic_cores_010kyr.jpg
      The pore diameter of ice is small enough to prevent the escape of CO2 out of the bubbles, but O2 and the smaller noble gases may escape at relaxation time, when the ice is expanding during a year or more at -20°C after drilling, which may give a small error in the O2/N2 ratio, but an important error in O2 isotope ratio. See:
      http://icebubbles.ucsd.edu/Publications/closeoff_EPSL.pdf

      • Ferdinand Engelbeen:
        Thank you.
        Interesting presentation that more or less confirms some of my questions; e.g. firn as being porous.
        I have some issues or concerns with the information in the presentation:

        Page 4, Slide 2: “Ice-age gas-age difference estimates from δ15N measurements
        “…GRIP, Greenland Dependence of linear correlation coefficient (r2) on 2 temperature proxies, δ18Oice and δ15N, on the mean Δage (7800-8780 y B.P.)”

        As I read that caption, confirmation CO2 levels are dependent and derived from temperature proxies.

        ” Page 6, Slide 2: “Advantages/disadvantages of measuring greenhouse gas concentrations in ice core gas bubbles”
        Disadvantages:
        Must determine gas-age ice-age difference in order to date samples. Δage varies from several to thousands of years.
        Uncertainty as high as >1,000 years in deeper records. Gas-age ice-age difference will vary with time (due to changes in T, accumulation rate).
        Gradual bubble close off during the transition from firn to ice leads to a smoothing of atmospheric concentrations over time.

        The presentation does not address air movement along layers. It is likely that even after vertical CO2 migration is sealed off, movement along layers may be possible; especially is the ice has flowed over shifting land features that cause the ice to flex.
        Both the presentation and PDF are based upon models using assumptions.
        I note the insistence that Antarctica is assumed free from volcanic influence. Isn’t that an odd assumption now that some 25 volcanoes are identified in West Antarctica?
        Using Google Earth one can recognize that the volcanoes in Antarctica are an extension of the ‘Ring of Fire’ volcanic chain surrounding the Pacific.

      • As I read that caption, confirmation CO2 levels are dependent and derived from temperature proxies.
        CO2 levels slightly increase in stagnant air near the bottom, about 1% over 40 years, because of its higher weight, compared to N2/O2. The needed correction is derived from the increase in the 15N/14N ratio, as 15N also is increased near the bottom compared to 14N because of its higher specific weight. That is only a relative small correction factor and doesn’t make that CO2 levels are dependent of temperature proxies.
        The same isotope ratio of 15N/14N in the gas phase and 18O/16O in the ice phase are used to estimate the difference in age of gas and ice. That is of interest to know the lag of CO2 after T changes, but doesn’t influence the CO2 levels either.
        The presentation does not address air movement along layers.
        Indeed, but that is not important as the air composition at the same horizontal layer is essentially the same. The horizontal sealing is mainly when all bubbles are fully closed and any ice movements keep the bubbles with the ice flow. The main disadvantage is that with ice flow (and depth) the ice layers are getting thinner and that you need more ice/layers to measure CO2.
        Most ice cores are drilled at summits, as there is less ice movement. Exceptions are Vostok and one of the Law Dome cores which were taken on slopes.
        Yes they used models, but they compared the models with reality, which showed that the models were not far off.
        Despite that all, the timing and age difference between gas and ice still are quite uncertain for the ice cores with small amounts of precipitation and long history. But the importance is only for establishing the lag of CO2 after T changes and doesn’t affect the CO2 measurements themselves.
        As far as I know, there is only one active volcano in Antarctica (Mt. Erebus) which is thousands of km from any drilling site. Which doesn’t mean that there can’t be an influence (not direct on CO2 levels), but that should be noted as a dust layer in the ice core record. Dust may influence CO2 levels in the ice if it is quite acidic (as is the case for Icelandic volcanoes) together with carbonate deposits from sea(salt) spray, but in Antarctic ice there is far less carbonate deposit, thus CO2 is less affected even with acidic dust.

  37. Another fly in the ointment (or firn) is described in this entry:
    http://wattsupwiththat.com/2012/06/16/younger-dryas-the-rest-of-the-story/
    Look for “bioturbation”, which smears annual changes into multi-annual averages. This is a problem for sediments unless the sediments are in anoxic areas. I would see an analogy to bioturbation at the very top, where blowing wind performs the same mixing the top layer of snow as biological mixing of the top layer of sediment.

  38. Naturally, CO2 outgasses at the surface of the ocean in response to warming to bring into equilibrium the partial pressure of the dissolved CO2 to the atmospheric concentration.
    Also, note that the chemistry texts overstate carbonic acid by a factor of 1000 to simplify the equilibrium equations with for the disassociation with carbonate and bicarbonate ions in sea water. Solving for three variables, ([H2CO3]+aq([CO2])), [HCO3], and [(CO3)2] is an equation with three unknowns, whereas solving for four unknowns is a bit more complicated.
    “Although the concentration of CO2(aq) far exceeds that of dissolved H2CO3 (in the order of 10³) we denote the concentration of all dissolved CO2 by [H2CO3]” from – from section 9.2 of CHEMISTRY OF CARBONIC ACID IN WATER.

    • The combination of CO2 (gas) and carbonic acid is only 1% at the pH of seawater. 90% is bicarbonate and 9% is carbonate. Doesn’t matter much if free CO2 and carbonic acid are lumped together.
      The free CO2 is what matters for the ocean surface – atmosphere equilibrium (Henry’s law) and of course the equilibrium reactions between the different carbon forms do change if free CO2 is removed from the oceans or reverse.
      What we know from experiments and real world observations is that the air-ocean equilibrium shifts with between 4 and 17 ppmv/°C according to different sources and the historical equilibrium as found in ice cores show a rather linear change of 8 ppmv/°C over the past 800,000 years.

  39. Jaworowski, one of the leading experts on ice cores and ice core data considers that 30–50% of CO2 is lost from ice cores during the traumatic process of extraction form the depths. If you back calculate the losses, CO2 ends up being the same or seriously higher than they are now. It is stupid and ignorant to consider that CO2 concentrations from ice cores are 100% absolute.

    • higley7 January 25, 2015 at 7:40 pm Edit

      Jaworowski, one of the leading experts on ice cores and ice core data …

      Jaworowski was nothing of the sort. He was a physicist whose expertise was in radiation. See my comment above, along with Ferdinand Engelbeen’s comments.
      Finally, think about the claim that “30–50% of CO2 is lost from ice cores during the traumatic process of extraction form the depths”. Lost by what means? How is that supposed to work? Is the CO2 in the air bubbles lost, but not the rest of the air in the air bubbles? Is the CO2 (at ~180 ppmv) supposed to mix with current air? That would make the CO2 measurement larger, not smaller. How could they “lose” CO2 but not air?
      I fear your claim just doesn’t make physical sense.
      w.

      • Willis Eschenbach
        I refer you to my refutation of your insults to the late, great ‘Zeb’ J. which I posted above here.
        Richard

      • Willis, you are walking on thin ice here. Stick to arguing Jaworowski’s ideas rather than questioning his expertise. How many times has your “expertise” been questioned…

      • Heavens Wills…lost to atmosphere. You have an untouched sample at depth at 100’s and to 1000’s of psi. You bring a core sampler to it. Long before you even get to it, long before you drill it, long before you extract it to the surface, but throughout all these, it’s been depressurized and is degassing to atmosphere. And preferentially doing so at the highest gradients…at the peaks of CO2 recordation.

      • Engineer Ron January 26, 2015 at 8:05 am Edit

        Heavens Wills…lost to atmosphere. You have an untouched sample at depth at 100’s and to 1000’s of psi. You bring a core sampler to it. Long before you even get to it, long before you drill it, long before you extract it to the surface, but throughout all these, it’s been depressurized and is degassing to atmosphere. And preferentially doing so at the highest gradients…at the peaks of CO2 recordation.

        Some of the bubbles are certainly lost to the atmosphere. But many others are not. In addition to cracking the ice core also flexes and expands. Remember, it is not being exploded. It is simply expanding back to its original uncompressed state. As a result, there are many, many bubbles still enclosed in the sample when it gets to the surface. Those bubbles are not released until the ice is shaved in the laboratory and their CO2 content is measured.
        So, does the ice crack as it comes up? Of course. Are there still untouched bubbles containing uncontaminated gas? You bet. Here’s a photo showing both things going on:
        https://wattsupwiththat.files.wordpress.com/2015/01/ice-core-section-with-air-bubbles.jpg
        Next, Ron, your explanation doesn’t explain why all of the ice core CO2 measurements are LOWER than current levels. Even if the cracking were to have replaced the air in every bubble with current air, we’d get current levels. What is your explanation for the lower levels found?
        Finally, why is there such good agreement between all of the ice cores? Look again at my graphic above. If your explanation were correct, we’d see answers all over the map, because the gas in the ice core bubbles would have been polluted by the modern higher levels.
        But we don’t see that. Instead, we see good agreement both in modern times (as I showed in my graphic) and in geological times.
        Regards,
        w.

      • The Pompous Git January 26, 2015 at 2:22 am Edit

        Willis, you are walking on thin ice here. Stick to arguing Jaworowski’s ideas rather than questioning his expertise. How many times has your “expertise” been questioned…

        Thanks, Git. I wasn’t making an “ad hominem” attack. Instead, I was merely responding to the “appeal to authority” fallacy inherent in the claim that Jaworowski was “one of the leading experts on ice cores and ice core data”.
        w.

    • @ higley7

      30–50% of CO2 is lost from ice cores during the traumatic process of extraction form the depths.

      I could agree with that estimate.
      Just the mechanics of “drilling” the ice core would relieve the horrendous lateral pressure on the core itself. Iffen ya drill into a confined “pocket” of NG … ya better have a good hold on the “drill stem”.
      And iffen those ice core are not completely sealed off before being lifted to the surface the decrease in air pressure would surely cause an outgassing of CO2.
      Those trapped CO2 “bubbles” would attempt to act like a weather balloon as it ascends into the high atmosphere.

      • Samuel, have you read any work on ice cores that confirms the 30-50% loss of CO2 from the bubbles, besides of what Jaworowski wrote? And why only a CO2 loss? If there are cracks during expansion of the ice at relaxation after drilling, then these will loose some compressed air, not alone CO2 – to the contrary – and ultimately get the same CO2 levels of the environment, not much lower levels as measured.

      • @ Ferdinand Engelbeen:January 26, 2015 at 2:18 pm

        If there are cracks during expansion of the ice at relaxation after drilling, then these will loose some compressed air, not alone CO2 – to the contrary – and ultimately get the same CO2 levels of the environment, not much lower levels as measured.

        Ferdinand, whatta yu trying to tell me, ….. that they measure the CO2 in those ice cores … BEFORE they drill and lift them out of the glacier?
        And just how do they perform that “trick”?

      • Samuel, Jaworowski is the man who said that 30-50% of all CO2 was lost by drilling and relaxation, not me. He doesn’t give one shred of evidence for his claim and as far as I know, he hasn’t done any work on CO2 in ice cores, not before, during or after drilling.
        All what I know from chemistry and diffusion gives the opposite answer of what Jaworowski said.
        Anyway, I am eager to learn: if you have any good arguments which show that Jaworowski may be right that CO2 is preferentially lost from the inside air via cracks into an outside atmosphere which is richer in CO2, I like to know that mechanism.

      • Ferdinand Engelbeen January 27, 2015 at 5:15 pm

        Anyway, I am eager to learn: if you have any good arguments which show

        Iffen I had had a mouth full of coffee when I read that I would have splurted it out all over my PC.
        Ferdinand, the only thing you are eager to do is “change-the-subject” when confronted by a naysayer of your commentary by injecting something “new, different & unrelated” to get your opponent off track ….. so that you can continue on “pushing your agenda”,
        And that’s a fact.
        I mean like, just what the ell does Jaworowski’s claims, credentials or expertise have to do with my above statement to you @ January 27, 2015 at 11:09 am?

  40. A lateral observation from this discussion; the dust layers in ice cores are accepted it seems by all, are caused by cold dry periods with higher winds, no doubt caused by a higher temperature gradient from tropics to poles with cooling.
    So how is it that climate change, global warming, weather wierding, are blamed for causing more severe weather events when tropic to poles temperature gradient is reducing by the purported warming of the poles?

    • I am pretty sure the “global weather wierding” only exists in the fantasy of climate models, which can’t even predict temperature, precipitation, cloud cover,… over a period of 20 years, let it be in 2050 or 2100…

      • Hi Ferdinand,
        I know it is late in the discussion, but could you look at the graph I posted pretty far up there and comment? It’s at:
        January 26, 2015 at 8:39 am
        I actually have come to trust your laying out the science here as you either really know what you are talking about or you are really, really good at faking it…;-)
        Some part of this you may have already answered. If (many of the) bubbles remain open to the atmosphere for ~30 to 40 years that might explain why Laws is in agreement with Mauna Loa, but only if somebody “cheated” in some way and selected bubbles that were closed and eliminated bubbles that gave the wrong answer, right? Basically what you are saying is that one can’t use a core less than 30 or 40 years old to normalize your proxy against Mauna Loa, because the bubbles are either a) still open and in equilibrium with the present or b) closed at some unknown time over the last 30-40 years and hence represent a non-uniform mix of concentrations depending on when EACH sampled bubble closed. Only when you get to where all of the bubbles are sealed to you get a consistent “record”, and even then at any given depth it is going to contain a mix of air that was finally sealed into bubbles across a range of years.
        I’d love your comments, however, on the illustrated variance and trend in the Laws (and Siple) data pre-1955, followed by the near-miraculous correspondence between Laws, Mauna Loa, and the last data point in Siple.
        rgb

      • RGB,
        Thanks, as always I enjoy your contributions as well thought and readable…
        I see that Willis already has done all the work…
        Some additions: Siple Dome measurements were done early 1980’s in the early times of ice core drilling, including the problems inherent with using new techniques. There were several parts of the core with cracks where drilling fluid was found and for small adjacent parts of the ice core enormous differences in CO2 levels. The researcher (Neftel) rejected the high readings and only retained the lowest readings. I should have rejected all samples of that part of the ice…
        Anyway the error margin of Siple is much higher: from a few ppmv to +/- 25 ppmv range for some parts of the ice core – and more where drilling fluid was found.
        The repeatability of Law Dome measurements for multiple samples at the same part of the core was 1.2 ppmv (1 sigma), a lot better…
        Etheridge measured 95 Law Dome samples, of which 11 results were rejected: 2 because of post drilling partly melt of the sample, the others for various problems with the equipment (vacuum line, grating, GC,…).
        The bubble closing process doesn’t start until the firn pressure is high enough to reach a density of 0.83-0.85 kg/dm3. That is at about 72 m depth for Law Dome. Then it takes another several years/meters to seal al bubbles. The average age difference between bubbles sealed at different depths is not that great, as migration (via residual pore diameter) is restricted with increased density. So no need to choose the right bubble, as anyway you need a lot of bubbles to have enough CO2 for the measurement…
        Thus indeed you need at least 40-50 years old ice, but that contains bubbles of average ~10 years older than the atmosphere. Thus a core drilled in 1990 has ice which is 40-50 years old but air of average 1980 and thus can compared to direct air measurements of 1980…
        For other ice cores like Vostok, the “newest” ice is already already thousands of years old and as the complete sealing also needs several thousands of years, the air it contains is average pre-industrial…
        The Law Dome sweetly moves into the Mauna Loa record without any manipulation: they simply used the measurements and normal corrections (for enhanced levels of the heavy isotopes and molecules near the bottom of stagnant air) and compared the results with the South Pole CO2 data. That confirmed the model they used to calculate the ice age – gas age difference, which is the most problematic part of any ice core. A small error in the difference would show up as too high or too low CO2 levels compared to Mauna Loa at the same “age”.

  41. It seems I’ve caused a bit of a firestorm by my description of Dr. Jaworowski as a “crackpot”, so let me retract that statement entirely, root and branch.
    A more accurate description would be that he was an eminent and well-respected scientist in his own field, which was that of radiation and its transport and health effects, but that his ideas about ice cores were demonstrably wrong. In addition, once the ideas were demonstrated to be wrong by other scientists, he clung to them nonetheless.
    As an example, to the end of his life Jaworowski claimed that there was no difference between the age of the ice and the age of the air enclosed in the bubbles in the ice. Now, our guest author has just given us an excellent explanation of exactly why the ice age is different than the air age, complete with well-drawn diagrams illustrating the physical processes involved.
    And other than from Jaworowski, there is little scientific disagreement on that point … but despite that, and despite the scientific work of Etheridge laying out the specific details of the process, Jaworowski refused to accept that the air trapped in the ice was not exactly the same age as the ice itself.
    Which is why I pay no attention to Jaworowski’s claims.
    Look, folks, there are lots of perfectly valid reasons to question the theory that CO2 is the secret master control knob on the temperatures. And there are even valid questions about the ice core data, like the somewhat circular nature of the calculation of the ice age/air age value … but Jaworowski’s claims aren’t among them. Not everyone who disagrees with the hypothetical “97% consensus” is right …
    w.

    • Sorry, continuing the ice core discussion from far above here because the discussion above is at too great a nesting for the reply button to appear:
      Source for the emission data? I assume that this is the anthropogenic emission data as opposed to all sources, as well. Also, what are you plotting from Laws — the raw points? I’m plotting the spline distributed in law2006.txt, which if anything should smooth the raw data. Offhand the spline doesn’t really seem to agree with your plot of the raw data, but of course I’ll have to overlay the two to understand that. In particular, the timing of the great depression, world war II, and the decrease in atmospheric CO_2 are off by at least a decade.
      This leads once again to fluctuation-dissipation and through this to the Bern model. As far as I can see, there is absolutely no way to feed the data you show into the Bern model and have decreasing CO_2 anywhere, ever come out. Unless, of course, there are time dependent changes in its parameters which are not in the model (that is, unless the model is incomplete). The extra CO_2 should have a residence time that is very, very long, and the system shouldn’t have been able to reach equilibrium, let alone decrease, while humans continued to add CO_2 across even the great depression.
      So one has to ask — (what) can one infer (about) the actual relaxation time(s) from some mix of Laws and the emission data? Also, what are the uncertainties on the emission data, and how is this estimated? There are some curiosities in this — for example, US production of coal peaked in 1944 at 683 million tons, a figure that was not exceeded until 1974 and over twice the production during the great depression. In Germany, over half of its gasoline and almost all of its aviation fuel were synthesized from coal at the same time the coal was used to drive its industrial production, the former at a horrible trade-off in total energy consumed (one has to burn the coal to electrolyze the water to make the hydrogen to bond to the carbon, all at high temperature and with major thermodynamic losses). This production continued from the late 30’s until the Allies bombed their synfuel plants in mid-late 1944 and 1945.
      So it’s not that I’m cynical or anything, but following your own requests at the head of all of your posts please cite your source for this so I can look at the data for myself.
      rgb

      • rgbatduke January 26, 2015 at 2:34 pm

        Sorry, continuing the ice core discussion from far above here because the discussion above is at too great a nesting for the reply button to appear:

        Yeah … I wish the format were wider with more nesting, OR no nesting.

        Source for the emission data? I assume that this is the anthropogenic emission data as opposed to all sources, as well.

        The emissions data is from CDIAC, and includes land use change, also from CDIAC. Hang on … OK, the emissions data is here, and the LULC data is here.

        Also, what are you plotting from Laws — the raw points? I’m plotting the spline distributed in law2006.txt, which if anything should smooth the raw data. Offhand the spline doesn’t really seem to agree with your plot of the raw data, but of course I’ll have to overlay the two to understand that. In particular, the timing of the great depression, world war II, and the decrease in atmospheric CO_2 are off by at least a decade.

        I’m using the data from ftp://ftp.ncdc.noaa.gov/pub/data/paleo/icecore/antarctica/law/law_co2.txt … which now appears to be a dead link … grrr, I hates it when they pull data like that. Looks like your data is newer than mine in any case.

        This leads once again to fluctuation-dissipation and through this to the Bern model. As far as I can see, there is absolutely no way to feed the data you show into the Bern model and have decreasing CO_2 anywhere, ever come out. Unless, of course, there are time dependent changes in its parameters which are not in the model (that is, unless the model is incomplete). The extra CO_2 should have a residence time that is very, very long, and the system shouldn’t have been able to reach equilibrium, let alone decrease, while humans continued to add CO_2 across even the great depression.

        Yeah, that’s one reason why I disagree with the Bern model, which seems extremely un-physical to me. Along with Jacobson, I put the e-folding time much shorter, at somewhere around 35 years. In any case, the ices cores taken as a total don’t seem to me to indicate a decrease in atmospheric CO2. Here’s my data again:
        https://wattsupwiththat.files.wordpress.com/2015/01/mauna-loa-and-ice-core-data-1850-2010.jpg
        I wouldn’t call the totality of the ice core data around 1945 a reduction.

        So one has to ask — (what) can one infer (about) the actual relaxation time(s) from some mix of Laws and the emission data? Also, what are the uncertainties on the emission data, and how is this estimated? There are some curiosities in this — for example, US production of coal peaked in 1944 at 683 million tons, a figure that was not exceeded until 1974 and over twice the production during the great depression. In Germany, over half of its gasoline and almost all of its aviation fuel were synthesized from coal at the same time the coal was used to drive its industrial production, the former at a horrible trade-off in total energy consumed (one has to burn the coal to electrolyze the water to make the hydrogen to bond to the carbon, all at high temperature and with major thermodynamic losses). This production continued from the late 30’s until the Allies bombed their synfuel plants in mid-late 1944 and 1945.

        Good question. The source document says:

        Publications containing historical energy statistics make it possible to estimate fossil fuel CO2 emissions back to 1751. Etemad et al. (1991) published a summary compilation that tabulates coal, brown coal, peat, and crude oil production by nation and year. Footnotes in the Etemad et al.(1991) publication extend the energy statistics time series back to 1751. Summary compilations of fossil fuel trade were published by Mitchell (1983, 1992, 1993, 1995). Mitchell’s work tabulates solid and liquid fuel imports and exports by nation and year. These pre-1950 production and trade data were digitized and CO2 emission calculations were made following the procedures discussed in Marland and Rotty (1984) and Boden et al. (1995). Further details on the contents and processing of the historical energy statistics are provided in Andres et al. (1999).
        The 1950 to present CO2 emission estimates are derived primarily from energy statistics published by the United Nations (2013), using the methods of Marland and Rotty (1984). The energy statistics were compiled primarily from annual questionnaires distributed by the U.N. Statistical Office and supplemented by official national statistical publications. As stated in the introduction of the Statistical Yearbook, “in a few cases, official sources are supplemented by other sources and estimates, where these have been subjected to professional scrutiny and debate and are consistent with other independent sources.” Data from the U.S. Department of Interior’s Geological Survey (USGS 2012) were used to estimate CO2 emitted during cement production. Values for emissions from gas flaring were derived primarily from U.N. data but were supplemented with data from the U.S. Department of Energy’s Energy Information Administration (1994), Rotty (1974), and data provided by G. Marland. Greater details about these methods are provided in Marland and Rotty (1984), Boden et al. (1995), and Andres et al. (1999).

        As to how accurate that makes it, well, it’s likely as good as anything else in climate, which is to say pretty decent but not great.

        So it’s not that I’m cynical or anything, but following your own requests at the head of all of your posts please cite your source for this so I can look at the data for myself.

        And rightly so, my friend, rightly so.
        As always, it’s a pleasure to have your comments.
        w.

      • In reference to graph @ Willis Eschenbach: January 26, 2015 at 6:41 pm

        I wouldn’t call the totality of the ice core data around 1945 a reduction.

        I am also curious about that dip (reduction) in CO2 …. circa 1938 thru 1945.
        Especially given the fact that there had to have been “tens of billions of tons” of sequestered CO2 emitted into the atmosphere as a result of all the bombings, burnings and “firestorms” that occurred in England, Western Europe, Eastern Europe, Russia and the Pacific theaters during said 7 years of War.
        To wit:

        Burning wood to fuel power stations can create as many harmful carbon emissions as burning coal, according to a government report.
        Ref: http://www.bbc.com/news/uk-28457104

    • Willis
      It’s clear that you still have not taken trouble to inform yourself about Z. Jaworoski.
      He was an expert on ice core analysis and a pioneer in that field and this includes CO2.
      That is why he was invited to testify before Congress on the subject of CO2 extraction from ice cores, which see.
      So, get up to speed on this distinguished scientist and quit saying that he was not knowledgeable in the subject of ice cores and CO2.

      • mpainter, I have been looking for his real expertise in ice cores, but didn’t find any. His expertise was clearly in radiation and health from radiation. There are some hints that he organized a few expeditions in the Arctic and Antarctic, but again it seems mainly to look for metal (ion) deposits and migration in ice, not about CO2.
        If you have more detailed information that he was an expert of CO2 in ice cores, I would like to see that cleared. As far as I can see now, his expertise on that subject was just theoretical and because of the many unphysical assumptions also completely wrong…

      • mpainter, I fear that whether a man testifies before Congress means nothing. I mean, James Hansen has testified before Congress, as has Al Gore … does that somehow convert them into “experts”?
        I have listed above some of Jaworowski’s more egregious mistakes. Ferdinand Engelbeen has listed others. I’ve asked you or anyone to show he was right regarding the issues we identified. To date, neither you nor anyone has done so.
        For example, as I showed above, Jaworowski doesn’t believe that there is a difference between ice age and air age. He called that idea a “lie”.
        Now, Ron Voisin in the head post gave us a complete explanation of the physical process whereby the ice age ends up being different from the air age, complete with illustrations … are you saying Voisin is wrong, and everyone else who has studied the ice age-air age question is wrong?
        Because that’s exactly what Jaworowski says, and that’s just one of his many errors. I discussed other of his errors here.
        So I fear that saying he was a “pioneer in the field” means nothing. I looked, and looked hard, for some examples of any independent research that he did on CO2 in ice cores. And I’ve asked for examples of such research. Nobody came up with any. Instead, I found two papers, one on lead in ice cores and one on radioactive materials in ice cores, but no original research on CO2 in ice cores. I did find a host of laughable claims from him that the methods used in ice core analysis were wrong … but no original research on how to do them right.
        But whether he did original research is not the issue. If you want to get any traction, what you need to do is show that his claims are true. So c’mon, mpainter, demonstrate to us that as Jaworowski claimed, there is no difference between ice age and air age, and the head post is wrong.
        Or you could take on this howler of his which I discussed above:

        … 97% of the total annual emission of CO2 into the atmosphere comes from natural emissions of the land and sea; human beings add a mere 3%. This man-made 3% of CO2 emissions is responsible for a tiny fraction of the total greenhouse effect, probably close to 0.12%. Propositions of changing, or rather destroying, the global energy system because of this tiny human contribution, in face of the large short-term and long-term natural fluctuations of atmospheric CO2, are utterly irresponsible.

        To which I replied:

        If you can’t see the huge fallacy in that argument, you need to contemplate the meaning of “equilibrium”. Here’s an example. If I make $100,000 per year and spend $100K per year, my net worth doesn’t change. My bank balance is at equilibrium.
        But if my expenses stay the same and I add “a mere 3%” to my income, my net worth will be 150% of the equilibrium value within a mere 17 years.

        Or as Ferdinand observed above about that same foolish claim of Jaworowski’s,

        Samuel, it seems quite difficult to understand that a small disturbance of a few % of what is a natural cycle may give a large increase (or decrease) over time. Something many housewives have no problem with: if they slightly spend more money than their household budget allows, they know they are getting into trouble…

        So point out to me how I’m wrong in that, and Ferdinand is wrong, and Jaworowski is right.
        The problem with Jaworowski is simple, and it has nothing to do with whether he is a nice guy, an eminent scientist, an expert on radioactivity, has testified before Congress, is a pioneer in the field, or anything else to do with his career or his personality.
        The problem with Jaworowski is simply that most of his claims about ice cores were clearly and demonstrably not true.
        w.

      • Oh, crap, Willis, the issue is his qualifications. Dispute his conclusions, fine I don’t give a hoot; I never commented on anything but his qualifications. So I will pass on your attempt to inveigle me in issues of science. I will only note that he is not the only qualified scientist who questions the reliability of ice core CO2 data.
        I note that you have the decency to withdraw your comment that he was a “crackpot” and I approve.

      • Dig deeper, Ferdinand. By his own profession, he organized about ten glacial expeditions to recover ice cores for study. He was one of the first to analyze ice deposits for pollution.
        I will stand on my original statement in this regard. Please read it again, carefully, and perhaps we can end this absurd discussion, thank you.

      • mpainter, I had read that he organized several expeditions to recover ice cores and test for pollution, but I couldn’t find anything about CO2. As far as I can see he is not an expert on CO2 measurements in ice cores, he only has written a lot of comments on the work of others in that field. Etheridge e.a. in 1996 responded to his criticism by carefully avoiding all pitfalls of drilling, clathrate formation, relaxation, sampling, etc. and additionally tested several assumptions used in firn densification models and migration directly in the field.
        That was published in 1996, but in 2004 and later, Jaworowski simply republished his earlier claims, refuted by the work of Etheridge, including in his letter to Congress.
        Thus no matter what his knowledge of ice cores was, his knowledge about CO2 in ice cores was completely outdated and completely wrong…

      • In reference to post January 26, 2015 at 10:30 pm

        …. the addition of 2.18 Gtonnes of C changes the atmospheric concentration by 1 ppmv … if not let me know. Note that this is 2.18 Gt of C that remains in the atmosphere, not 2.18 Gt emitted.

        My simple math calculation when atmospheric CO2 was at 393 ppm …. resulted in the fact that …. 1 ppm of atmospheric CO2 is equal to 5 Gts, ….. to wit:
        The average mass of the atmosphere is about 5 quadrillion (5,000,000,000,000,000) metric tons.
        The Carbon dioxide (CO2) in the atmosphere is currently at 393 ppm — or 0.0393%
        Thus, @ 0.0393%, there is about 1,965,000,000,000 metric tons of CO2 in the atmosphere.
        Or, there is about 1,965 billion tons of CO2 in the atmosphere.
        Therefore, @ 393 ppm, each ppm of CO2 is equal to about 5 billion tons of CO2.
        But now, just the weight of the carbon (C) in the CO2 would be 1.36 Gt per 1 ppm of atmospheric CO2 …… to wit:
        CO2 = 44. … (2 O2 at 16 and 1 C at 12) … and weight of 1 C is 27% of the CO2 weight.

      • Bart
        temperature phase does not match CO2 phase – it leads it by 90 deg. Those [two] conditions mean that your interpretation is wrong.
        Of course CO2 does not instantly follow T changes, it needs time. For a sinusoid change in temperature, CO2 follows with a 90 deg. lag. If you take the derivatives, both dT/dt and dCO2/dt shift 90 deg. back in time and if you look at the variability of T, that is perfectly synchronized with dCO2/dt, as good as with the opposite swings of the derivative from δ13C. The perfect match is the result of the use of the derivative variability of one of the variables and the direct variability of the other one. Plus an arbitrary factor and offset to match the slopes. But that has no bearing in any physical process: CO2 increase/decrease follows T increase/decrease with a 90 deg. lag and so do the derivatives…
        The derivative of T has near zero slope and the full variability of T. If you integrate the derivative to obtain its influence on the derivative of CO2, there is a perfect match in timing and near zero slope. The slope is from a different process which has nothing to do with T.

      • The average mass of the atmosphere is about 5 quadrillion (5,000,000,000,000,000) metric tons.

        You are working way too hard and need to pay attention to significant figures.
        Let’s make it easy: The increase in atmospheric CO_2 concentration upon the addition of 1 metric gigaton of CO_2 is:
        \Delta C_{CO_2} = 1\times 10^{9} / 5. \times 10^{15} = 2. \times 10^{-7}
        If we divide this by 10^{-6} to put it in parts per million, it is 0.2 ppm/GT of CO_2. Note well that I’m using long (metric) tons (1000 kg) throughout, since the mass is given that way, and that they are roughly 10% larger than an English/short ton.
        There is no point in giving more than one digit in the answer unless you plan to use 5.1 or 5.15 quadrillion metric tons throughout (with more digits). This way the answer carries its own statement of likely precision.
        If we then look WAY uphill at Willis’ graph of CO_2 input, we see that we are currently adding roughly 10 GT/year, which means that would be increasing CO_2 by roughly 2 ppm/year, IF 100% of the addition went straight into the atmosphere. Interestingly, this is almost exactly the current slope of the Mauna Loa data, suggesting that in fact pretty much 100% of our additions are indeed going straight into the atmosphere.
        This all by itself is rather curious, given the active CO_2 cycle and the system of participating reservoirs. Each additional ton seems to simply shift the equilibrium. This is actually evidence in FAVOR of a Bern-like model with very slow (if any) natural sequestration, a closed system of capacitors with no way to fundamentally bleed off additional charge (to use Bart’s favorite analogy). But the correspondence of the rate estimate assuming 100% retention to the measured slope is rather compelling…
        rgb

      • rgbatduke January 29, 2015 at 7:37 am

        The average mass of the atmosphere is about 5 quadrillion (5,000,000,000,000,000) metric tons.
        You are working way too hard and need to pay attention to significant figures.
        Let’s make it easy: The increase in atmospheric CO_2 concentration upon the addition of 1 metric gigaton of CO_2 is:
        \Delta C_{CO_2} = 1\times 10^{9} / 5. \times 10^{15} = 2. \times 10^{-7}
        If we divide this by 10^{-6} to put it in parts per million, it is 0.2 ppm/GT of CO_2

        Robert, I always hesitate greatly to disagree with you, having been wrong more than once … but your method gives an answer of 1.4 Gt C per ppmv, which is only about 2/3 of the actual figure. I think that like Samuel Cogar, you’ve left out the step of converting from parts per million by mass to parts per million by volume.
        And as regards significant figures, I was taught to leave in every swinging decimal during the calculation, and only cut it to significant figures once the work was done.
        In any case, always good to hear from you.
        w.

    • RGB,
      No need to use the Bern model, which is anyway not what reality says. The current decay rate (~51 years e-fold decay rate) still is the same as of 20 years ago, there is no sign of saturation of the deep oceans, as the Bern model supposes. An even simpler approximation is that about 53% of all human CO2 remains in the atmosphere:
      http://www.ferdinand-engelbeen.be/klimaat/klim_img/acc_co2_1900_cur.jpg
      I have the emission data somewhere, but I suppose that Willis has a better memory (and/or a better administration)…

      • Ferdinand, I’d never thought about graphing it that exact way, but that is a marvelous graph. It was such graphs (in my own more complex form) that originally convinced me that the increase in CO2 is of human origin.
        w.

      • And yet again, how do you compute cumulative emissions. What ARE cumulative emissions ppmv? Mass of emissions over mass of the atmosphere? How, in other words, would I reproduce this figure from the ice core data and e.g. Willis’ CDIAC data?

      • Robert, I think the number you are looking for is that the addition of 2.18 Gtonnes of C changes the atmospheric concentration by 1 ppmv … if not let me know. Note that this is 2.18 Gt of C that remains in the atmosphere, not 2.18 Gt emitted.
        w.

      • RGB,
        Cumulative emissions are the sum of all human emissions translated from GtC fossil fuel use to ppmv (factor 2.12 GtC/ppmv). If there was no removal by the increased CO2 partial pressure, that would stay in the atmosphere forever. But the increased pressure pushes more CO2 into the oceans at the cold polar sink places (and in plant alveoli) and suppresses the release of CO2 at the upwelling places near the equator. Thus some of the extra CO2 is removed. For the current extra CO2 (110 ppmv) above the temperature driven equilibrium, and emissions – increase in the atmosphere is about 2.15 ppmv/year that gives:
        110 ppmv / 2.15 ppmv/year = ~51 years e-fold decay rate for a linear process.
        Which it seems to be: the uptake and release of CO2 by the oceans is quite linearly in ratio to the partial pressure difference between atmosphere and ocean surface and over ice ages, there is a quite linear ratio between temperature and CO2 levels of ~8 ppmv/K.
        Peter Dietze had a similar e-fold rate years ago:
        http://www.john-daly.com/carbon.htm
        That the increase/emissions ratio is still the same over the past 110 years is not coincidence: human emissions increased slightly quadratic over time, which led to a slightly quadratic increase in the atmosphere and thus a slightly quadratic increase in the atmosphere and thus a rather fixed ratio between increase in the atmosphere and increase of human emissions:
        http://www.ferdinand-engelbeen.be/klimaat/klim_img/temp_co2_acc_1900_2011.jpg
        I only forgot to include the offset of emissions and increase in 1900…
        Over short periods (2-3 years) there are disturbances of a few ppmv caused by short term temperature changes, mainly a fast reaction of vegetation on temperature and drought (El Niño). After a few years, that levels off and the increase in the atmosphere simply follows human emissions:
        http://www.ferdinand-engelbeen.be/klimaat/klim_img/dco2_em4.jpg
        where the red line is the calculated increase from the emissions minus the sink rate as function of the pCO2 difference between CO2 in the atmosphere and the equilibrium setpoint with baseline 295 ppm in 1960 and 8 ppmv/K change in setpoint with an e-fold decay rate of 51 years…

      • @ Ferdinand Engelbeen January 26, 2015 at 3:51 pm

        An even simpler approximation is that about 53% of all human CO2 remains in the atmosphere:

        Ferdinand, just what sort of a sneaky trick are you using to correlate said 53% of all yearly human CO2 emissions from 1900 to 2011 that remain residual in the atmosphere …. as denoted by your “CumEmiss – CO2 increases 1900 -2011” graph ….. with the exponential increase in world population of humans … and/or … the steady and consistent increase in measured atmospheric CO2 ppm …. from 1900 to 2010, to wit:
        year — world popul. – % incr/Decade – CO2 ppm – % incr. — avg increase/year
        1900 – 1,564,000,000 est. ________ 280 ppm est.
        1940 – 2,300,000,000 est. ___ ____ 300 ppm est.
        1950 – 2,556,000,053 – 11.1% ____ 310 ppm – 3.1% —— 1.0 ppm/year
        1960 – 3,039,451,023 – 18.9% ____ 316 ppm – 3.2% —— 0.6 ppm/year
        1970 – 3,706,618,163 – 21.9% ____ 325 ppm – 2.7% —— 0.9 ppm/year
        1980 – 4,453,831,714 – 20.1% ____ 338 ppm – 3.8% —– 1.3 ppm/year
        1990 – 5,278,639,789 – 18.5% ____ 354 ppm – 4.5% —– 1.6 ppm/year
        2000 – 6,082,966,429 – 15.2% ____ 369 ppm – 4.3% —– 1.5 ppm/year
        2010 – 6,809,972,000 – 11.9% ____ 389 ppm – 5.1% —– 2.0 ppm/year
        2012 – 7,057,075,000 – 3.62% ____ 394 ppm – 1.3% —– 2.5 ppm/year

      • OOPS, the following should have been posted in this thread.
        ===========
        In reference to post January 26, 2015 at 10:30 pm

        …. the addition of 2.18 Gtonnes of C changes the atmospheric concentration by 1 ppmv … if not let me know. Note that this is 2.18 Gt of C that remains in the atmosphere, not 2.18 Gt emitted.

        My simple math calculation when atmospheric CO2 was at 393 ppm …. resulted in the fact that …. 1 ppm of atmospheric CO2 is equal to 5 Gts, ….. to wit:
        The average mass of the atmosphere is about 5 quadrillion (5,000,000,000,000,000) metric tons.
        The Carbon dioxide (CO2) in the atmosphere is currently at 393 ppm — or 0.0393%
        Thus, @ 0.0393%, there is about 1,965,000,000,000 metric tons of CO2 in the atmosphere.
        Or, there is about 1,965 billion tons of CO2 in the atmosphere.
        Therefore, @ 393 ppm, each ppm of CO2 is equal to about 5 billion tons of CO2.
        But now, just the weight of the carbon (C) in the CO2 would be 1.36 Gt per 1 ppm of atmospheric CO2 …… to wit:
        CO2 = 44. … (2 O2 at 16 and 1 C at 12) … and weight of 1 C is 27% of the CO2 weight.

      • Samuel,
        I am pretty sure that you will find a correlation between the increase of the world population and its increasing wealth and the increase in CO2: more people use more energy, thus both go up more or less exponential (quadratic in this case).
        You can plot the year by year emissions and increase in the atmosphere and what that means for the sink rate:
        http://www.ferdinand-engelbeen.be/klimaat/klim_img/dco2_em2.jpg
        In the period 1959-2011 human emissions increased a 4-fold, the increase in the atmosphere increased a 4-fold and the sink rate increased a 4-fold.
        There is a lot of (temperature caused) noise around these trends, but the trends are clear: each a 4-fold increase over a period of 52 years, directly caused by the human emissions…

      • Bart, I have repeatedly said to you that your graph gives a false impression of problems between the rate of change in CO2 and human emissions: you use different units, scales and an offset for one of them, while the conversion of units can be easily done. After five times of protest from me, that is a deliberate lie.
        The reality is that human emissions are average twice the increase in the atmosphere and that nature is a net sink for CO2: temperature variations are responsible for the variations in sink rate not for the increase in the atmosphere. That can be seen in my graph where emissions, rate of change in the atmosphere and sink rate are all plotted on the same scale and units…
        There is a marvelous match between the variability of temperature and the CO2 rate of change, but no causal connection between temperature and the CO2 increase of 110 ppmv in the atmosphere as that violates Henry’s law of solubility in seawater and there is not a shred of evidence of increased outgassing of the oceans or an increased throughput of CO2 through the atmosphere, to the contrary: the latest estimates of the residence time of a CO2 molecule in the atmosphere slightly increased over time…

      • “After five times of protest from me, that is a deliberate lie.”
        Frankly, Ferdinand, your protests are dumb. If you don’t like the scale, all you have to do is cover up the right hand scale and not look at it. The plot itself is an affine map. There is nothing wrong with it.
        “The reality is that human emissions are average twice the increase in the atmosphere and that nature is a net sink for CO2…”
        Which I have repeatedly told you is immaterial to the question of attribution – all that matters is the power of the sinks. After all my protests, this is a deliberate lie.
        “…but no causal connection between …”
        If CO2 enriched waters are upwelling, then there is a causal connection. We do not avert our eyes from an oncoming train just because its not on schedule. The reality is, there is a heavy mass moving towards you and you better get out of the way. Just so, when you see the effects of a mechanism clearly evident in the data, you do not demure from investigating it because you do not, at the moment, know for certain what caused it. You investigate further, and find the cause.
        “…there is not a shred of evidence of increased outgassing of the oceans or an increased throughput of CO2 through the atmosphere…”
        YES THERE IS!!! The data that show that the rate of change of CO2 is due to temperature is that evidence.

      • “The plot itself is an affine map. There is nothing wrong with it.”
        Furthermore, it shows more clearly what you are trying to cover up, which is that emissions are accelerating while concentration is not. That is a FACT, Ferdinand.
        Emissions are accelerating while rate of change of concentration is flat, and temperature phase does not match CO2 phase – it leads it by 90 deg. Those two conditions mean that your interpretation is wrong.

      • Bart, your plot is a nice example of what I have read in a book titled “How to Lie with Statistics”. If you use different scales and an offset, you can suggest any problem that doesn’t exist or exaggerate small changes into disasters. If plotted on the same scales without any offset, the increase in the atmosphere still is widely within the natural variability of the sink rates. The rate of increase in the atmosphere was even negative with increasing CO2 emissions in the period 1976-1993 as can be seen in the second graph here. Despite the natural variations in sink rate, the overall increase rate is directly in ratio to the increase in partial pressure of CO2 in the atmosphere, compared to the partial pressure of CO2 in the oceans for the actual temperature.
        YES THERE IS!!! The data that show that the rate of change of CO2 is due to temperature is that evidence.
        That just proves the opposite, as the variability in CO2 rate of change is entirely caused by the influence of temperature variability (El Niño) on (tropical) vegetation, not on the oceans, as the opposite δ13C and CO2 changes in rate of change show:
        http://www.ferdinand-engelbeen.be/klimaat/klim_img/temp_dco2_d13C_mlo.jpg
        As the longer term changes in vegetation shows more CO2 uptake than release (based on the oxygen balance), vegetation is not the cause of the increase of CO2 in the atmosphere. The earth is greening.
        Again, there is not a shred of evidence that the oceans increased in output/throughput, as the more recent estimates show an increase, not a decrease in residence time for CO2 and the oceans are a net, increasing sink for the extra CO2 in the atmosphere:
        http://www.pmel.noaa.gov/pubs/outstand/feel2331/mean.shtml

      • You don’t get it. Keep watching. Human emissions keep accelerating. CO2 maintains a steady space. Temperatures lead CO2 by 90 deg. There is no doubt about it. This rise is almost entirely natural.

  42. Willis and Ferdi,
    CO2 (and O2/N2) are dissolved in the water become ice. Some is trapped as bubbles, much is not. Pressurization takes place slowly for 100’s or 1000’s of years. Then later you have an untouched sample at depth at 100’s and to 1000’s of psi. You bring a core sampler to it. Long before you even get to it, long before you drill it, long before you extract it to the surface, but throughout all these, it’s been depressurized and is degassing to atmosphere. Any and all components of the sample that can volatilize from this depressurization will volatilize. One of these components is CO2. It varies in concentration sample to sample and within a sample (the sought information). Entropy being what it is there will be homogenization. And preferential outgassing of CO2 at the highest gradients…at the peaks of CO2 recordation. But this is at the back-end.
    At the outset, as soon as the then current atmospheric CO2 concentration began to fall from previously being higher, the whole snow/firn/ice stack began outgassing to atmosphere throughout this concentration falling phase. When we examine Mauna Loa today and compare to recently trapped ice-core CO2, we find good agreement. But this false-positive confirmation exists because the outgassing phase has yet to begin.
    Then you have the temporal sampling issue further diminishing the peaks.
    Try as you might to imagine that these are “small corruptions”, they are not. Ferdi is want to linearize this to say if X-attenuation happened 100kya then why don’t we see 2X-attenuation at 200kya. But all the deglaciation peaks received similar attenuation.
    Natural CO2 sourcing is a strong function of temperature. Natural CO2 sinking is a closely following strong function of CO2 concentration. Despite Ferdi’s want to linearize, sources can never get very far out in front as the further they do the more they stimulate the sinks. CO2 spiking occurs with every transition warming-to-cooling. We saw this in the chemical record. We see this plainly in the 800ky ice-cores. We see this plainly in the seasonal variation of Mauna Loa. We see this today with Earth beginning to cool.
    There, that’s a preview of the follow-up essay.

    • Engineer Ron: “At the outset, as soon as the then current atmospheric CO2 concentration began to fall from previously being higher, the whole snow/firn/ice stack began outgassing to atmosphere throughout this concentration falling phase.”
      To me you seem to be saying that the diffusion coefficient decreases with depth. If so, I assume the follow-up essay will address why this is so.

    • I mean yes. The diffusion coefficient is certainly a function of depth. But so long as atmospheric CO2 is climbing the firn will reasonably accurately record it. When atmospheric CO2 begins to fall the previous recordation is free to vent to atmosphere.

    • Ron, please, do read some works about ice core drilling, what you write here has no bearing in reality…
      1. There is no liquid water in snow/ice below -32°C, except where impurities are found. Snowflakes are formed directly from water vapor at about -80°C in Antarctica. Thus there is no dissolved CO2 or N2 or O2 in the snow/firn/ice of Vostok and Dome C and other cold inland ice cores.
      2. During drilling, especially for the deep cores of central Antarctica, drilling fluid is used (a mixture of petroleum and Freon) with the same density as the ice to prevent a collapse of the bore hole. The expansion of the ice core then is mainly at the surface where the ice is stored for one year or more at -20°C. During that period, the ice volume increases with about 50%.
      3. Degassing to the atmosphere? Maybe if there are cracks between the bubbles and the atmosphere. But that doesn’t make lower CO2 levels, that makes higher CO2 levels: CO2 has a higher affinity for water than O2 or N2. At -20°C, there is a small liquid-like water layer at the surface of the ice, where small amounts of CO2 may hide. Thus if there is an “escape”, it will be O2/N2 first, CO2 second and you will find higher levels of CO2, not lower.
      4. If there are cracks and the pressures equalize, the migration will take over and goes from the outside 400 ppmv (or higher in the laboratories with a lot of people walking around) to the inside 180-300 ppmv. Again that gives higher CO2 levels than original.
      5. There is no trace of migration in the cold inland ice cores themselves. If that was the case, levels during glacial periods would be too low to sustain (plant) life, endangering all life on earth. When the temperatures in the past got down, CH4 followed immediately, while CO2 followed with a long lag. If that was a matter of migration, CH4 would give more troubles than CO2.
      6. If the CO2 levels would go down, the levels down the firn will simply follow and the ice core will also follow, be it smoothed, depending of the accumulation and temperature.
      7. If there is attenuation from migration within the ice itself, that doesn’t stop after 100,000 years, neither does it stop at 300 ppmv. It only stops if there are no concentration differences anymore. Thus time is crucial here.
      8. CO2 sourcing is a matter of temperature: 4-17 ppmv increase for 1°C temperature increase of the ocean surface. But an increase of temperature in general increases plant growth, so the average is 8 ppmv/°C. Short term: 5 ppmv/°C over the seasons, 4-5 ppmv/°C over 2-3 years variations (Pinatubo, El Niño).
      How warm need the oceans to be to reach 110 ppmv extra?
      9. The current e-fold decay rate for the extra CO2 in the atmosphere is ~51 years. Fast enough to follow ice ages and reversals, but too slow to remove all human emissions on short note…

  43. We saw this in the chemical record: Obviously these scientists were dufus. They never noticed a time-of-day dependence of their measurements. They never noticed a spatial location dependence of their measurements. And they never noticed a seasonal dependence of their measurements. Despite their utter incompetence, most often they got the right answers. But here is the real comedy of their dufusness. In spite of them being strewn the world over, they coherently made systematic mistakes (for years) only during the years following temperature transitions of warming-to-cooling. During these periods they coherently and systematically over-estimated the then current CO2 concentration.
    Now hows that for being astrology-like – collective scientific corruption at its finest. Throw all that sh*t out and go with the itsy-titsy new infallible ice-core stuff.
    The new itsy-titsy stuff that is utterly incapable of resolving a hundred year long 100% perturbation at its outset. Or any perturbation that so surely accompanied the myriad transitions throughout the Holocene.

    • Ron, if you are talking about the historical CO2 measurements with wet chemical methods, then they couldn’t know that there were seasonal changes, as most equipment had an accuracy of +/- 10 ppmv (some were worse: +/- 150 ppmv!). A lot of measurements were taken in towns, forests, agriculture,… many were taken to look at the changes in CO2 level within growing crops, etc… All interesting, but not suitable to have an idea of what really happened in the bulk of the atmosphere. At that time, nobody did know that there was a “background” CO2 level in 95% of the atmosphere…
      It was the rigorous work of Keeling Sr. which showed that the huge day/night swings were caused by vegetation as he measured the 13C/12C ratio on the same CO2 samples and who noticed that afternoon levels were more equal everywhere and quite stable in deserts, away from direct sources and sinks.
      If you look only at the historical data which were taken on sea ships and coastal with wind from the sea, these are (widely) around the high resolution ice core data for the same period of the past…

  44. And why then did the ice-cores record CO2 spikes at the deglaciation intervals? Because these were enormous spikes (likely several hundred % perturbation) of enormous duration (1000’s to tens of 1000’s of years).
    My thinking of the Big-Picture – not so complicated.

  45. Coming into this 2 days late, so this may not be read…
    In addition to the gradients Ronald Voisin discusses there is this:
    In March 2004, Polish Professor Zbigniew Jawaworski made a written statement to the U.S. Senate Committee on Commerce, Science, and Transportation:

    False Low Pre-industrial CO2 in the Atmosphere
    Determinations of CO2 in polar ice cores are commonly used for estimations of the pre-industrial CO2 atmospheric levels. Perusal of these determinations convinced me that glaciological studies are not able to provide a reliable reconstruction of CO2 concentrations in the ancient atmosphere. This is because the ice cores do not fulfill the essential closed system criteria. One of them is a lack of liquid water in ice, which could dramatically change the chemical composition the air bubbles trapped between the ice crystals. This criterion, is not met, as even the coldest Antarctic ice (down to -73°C) contains liquid water[2]. More than 20 physico-chemical processes, mostly related to the presence of liquid water, contribute to the alteration of the original chemical composition of the air inclusions in polar ice[3].
    One of these processes is formation of gas hydrates or clathrates. In the highly compressed deep ice all air bubbles disappear, as under the influence of pressure the gases change into the solid clathrates, which are tiny crystals formed by interaction of gas with water molecules. Drilling decompresses cores excavated from deep ice, and contaminates them with the drilling fluid filling the borehole. Decompression leads to dense horizontal cracking of cores, by a well known sheeting process. After decompression of the ice cores, the solid clathrates decompose into a gas form, exploding in the process as if they were microscopic grenades. In the bubble-free ice the explosions form a new gas cavities and new cracks[4]. Through these cracks, and cracks formed by sheeting, a part of gas escapes first into the drilling liquid which fills the borehole, and then at the surface to the atmospheric air. Particular gases, CO2, O2 and N2 trapped in the deep cold ice start to form clathrates, and leave the air bubbles, at different pressures and depth. At the ice temperature of -15°C dissociation pressure for N2 is about 100 bars, for O2 75 bars, and for CO2 5 bars. Formation of CO2 clathrates starts in the ice sheets at about 200 meter depth, and that of O2 and N2 at 600 to 1000 meters. This leads to depletion of CO2 in the gas trapped in the ice sheets. This is why the records of CO2 concentration in the gas inclusions from deep polar ice show the values lower than in the contemporary atmosphere, even for the epochs when the global surface temperature was higher than now.

    Jawaworski worked for 40 years on glacier studies and organized 11 glacier expeditions.
    The above is from http://www.john-daly.com/zjiceco2.htm

    • Steve, you may have missed some of the discussion:
      The late Prof. Jaworowski was an expert on radiation and its dangers. Somewhere was a hint that he organized 11 expeditions to Arctic and Antarctic ice sheets. And he published two works about the migration of metal (ions) and radioactive fallout in ice cores. But…
      There is zero indication that he has the slightest knowledge of the behavior of CO2 in ice cores.
      To the contrary, the “arbitrary shift” of ice core data to match the Mauna Loa data as mentioned in his letter to Congress is pure nonsense: it is based on his own completely refuted idea (definitively in 1996 and even in the work he sites!) that there is no difference between the age of the ice and the enclosed bubbles.
      See: http://www.ferdinand-engelbeen.be/klimaat/jaworowski.html#The_arbitrary_shift_of_airice_data
      What he said on several items is so far off the mark, refuted by real field measurements and physically impossible, that the only conclusion one can draw is that his knowledge of ice core CO2 was non-existing.

  46. Ferdinand Englebeen: “his knowledge about CO2 in ice cores was completely outdated and completely wrong”
    ####
    Such an extravagant statement. He published in 2005 with citations current to that date. He evidently followed the subject at that time.

    • mpainter,
      which of his statements was current in 2005? He simply repeated his ideas of 1992, which were already completely refuted at the time he wrote them, like his idea that there is no difference between the average age of the gas bubbles and of the surrounding ice. Or the “leaking” of CO2 from 180 ppmv inside the ice cores towards 400 ppmv at the outside…

    • mpainter,
      What Jaworowski wrote in 2004, 2005 and later is simply a repeat of all his arguments from 1992, which were already wrong at that time. Completely ignoring the very detailed work of Etheridge e.a. of 1996 on three ice cores. Completely ignoring the work of other ice core specialists since 1996 a.o. the development of the sublimation technique to recover all CO2 from the ice cores, including from clathrates,…

  47. My own take on this comes from my design experience with 16-position hot runner plastic multiple-layer co-injection blow-molding machines, working in R&D in dealing with laminar flows. (Note the term “hot runner”; this is FAR different from 90% of all plastics, which have COLD runners (or passages).)
    When fluid materials with whatever viscosity (from runny to thick ooze) flows “laminarly” within or on solid surfaces, the fluid does not all flow at the same rate. Laminar flow is slow enough so that their is no turbulence, so that the layers STAY in layers. Mixing layers in our case was a VERY undesirable condition.
    The fluid right up against the solid surface has friction WITH that surface, and that friction make the flow slower than for the fluid farthest from the surface. The fluid in contact with the wall and for a very short distance away from the all is called the “boundary layer”. In between the boundary layer and the farthest fluid layer there is a flow gradient. In multi-layer plastics, the velocities of these layers is critical to getting each and every layer to arrive inside the plastic molds at the proper time. (This is necessary because the plastic bakes while in the heated passages, changing its properties, which is not good at all. So it is imperative to have equal time for all the layers in the passages/runners.)
    So I have experience dealing with layers within laminar fluid flows.
    One thing to be aware of is that in laminar flow the layers do NOT all flow the same speed. For glaciers, this is all important. But it does not ever seem to be taken into account in ice cores.
    Ice in glaciers – according to the experts – also is a laminar flow. Those layers are all moving at different speeds, down the mountain valley. The top surface is traveling at one speed, and each layer down moves a little slower – or perhaps faster in some cases, because of the water at the basal ice (that I think doesn’t always exist).
    The ice does not flow monolithically.
    This is actually known, that the ice moves in layers that are in shear – sliding over each other.
    So, how does this affect ice cores? The mere existence of these shearing layers means that an ice core is NOT measuring the ice laid down at that point throughout the history of the drilled ice core. If the layers are going different speeds, then as we look down an ice core, we need to understand the time element in the different flow rates. But I have never seen this discussed – ever.
    Example: If the ice 10 meters down traveled slower/faster than the ice at the bottom of the FIRN, then the history shown in the ice core at -10 meters is from a different time than the ice that was under that drilling location 10 years earlier. That ice may have been 2 meters or 20 meters up or downhill 10 years earlier.
    So, the ice core is reading ice laid down at MANY locations, all of which just HAPPEN to be in that vertical column at the time the drilling is done. The layers SEEN just happen to line up that way because the drilling was done THEN. If done a year later, the ice layers will be all different.
    NOW, add in this:
    The shearing of the layers screws up the supposedly “trapped” gases, churning them all the time. ALL THE TIME. One layer sliding over the other one carries material from the bottom of one over the next layer down. What effect does this have on the gases? One thing is certain: It does not leave those gases unaffected.
    Now, add in this, too:
    No two glaciers move at the same speed, nor over identical ground. Therefore the shearing internally between the layers in one glacier will be different from other glaciers.
    All of this is terrifically complicating the simplistic idea of the ice cores as a record of the past atmospheric gases.
    The simplistic idea cannot be correct.
    Until they have a way of quantifying and empirically testing the variables – and then keep on verifying that those variables are known and included – the entire trapped gases principle cannot do what they say it does. They can’t even tell if the ice in a layer was laid down at the same location as the layers above it. LITERALLY, the ice in any layer may have come from 1 or 2 or 3 km away from the top ice. Perhaps more.

    • Steve,
      Ice core specialists are not stupid. Most cores were drilled at the summits of domes of ice, see the names of the cores: Law Dome, Siple Dome, Taylor Dome, Dome C… Your shear argument doesn’t count there.
      There are a few exceptions like Vostok and the third ice core at Law Dome, which were taken downslope.
      These are corrected with depth for the different speeds of the ice, of course with a (large) margins of error, but not far off as the 420 kyear old lowest layers of Vostok fit the timing for the temperature proxies (dD and d18O) and CO2 of the middle layers of Dome C (which reach 800 kyear). The same for the downslope Law Dome ice core, where the ice layers can be compared to a lot of other ice cores over the past 1,000 years.

  48. Ferdi,
    You’re entitled to your opinions. I have mine.
    I’m right and you’re wrong. I’ll bet you a beer.
    The currently observed spike is, at the least, 95% natural and, at the most, 5% anthropogenic.
    Assuming the Earth temp falls a couple tenths in the coming few years atmospheric CO2 will spike hard; as the majority natural sinks respond to the lowering temp appreciable faster than the majority natural sources.
    We will endure cockamamie explanation like: Ma Nature is feed up; she cannot choke down our poison any more; hence the enhanced spiking.
    It will nonetheless get little traction as we go on to concern ourselves with the coming cold climate.
    Over time the CO2 will crash down but I cannot well predict the time lag. Shouldn’t take more than a decade or two.

    • Ron, if humans add 400 GtC as CO2 directly into the atmosphere and the measured increase is 232 GtC where does the difference go? Space?
      Maybe you know of an accountant who only shows your income, but forget to mention your expenses. If 95% input is natural and 5% is human but the output is 97.5% all natural, leaving an increase of 2.5% in the atmosphere, one need much better arguments to show that the increase is NOT caused by the human emissions.
      Further have a look at the period 1945-1975: temperature got (slightly) down, CO2 got up in perfect harmony with human emissions, only disturbed a little by temperature variations…

      • Into the natural sinks, mineral, biological, and others. If, as you hold, 168 GtC went into the sinks, how do you know the rest of it didn’t, too?
        “…CO2 got up in perfect harmony with human emissions…”
        No, it went up in perfect harmony with the integral of temperature anomaly. The fact that emissions and concentration both went up is not remarkable. Emissions going up is a given. Concentration could then be going up, or it could be going down. It’s just a 50/50 happenstance that they both went up.
        But, the rate of change of concentration was bobbling around, unlike the emissions, and both the bobble and the long term match the temperature data. The odds of that being happenstance are vanishingly small.

      • “long term match the temperature data”

        The past 15/16/17 years worth of temperature data is flat.
        CO2 went up over 30 ppm.

        Better get a new theory.

      • Says the Liberal Arts Walter Mitty who fancies himself a modern day Socrates, and doesn’t even understand the concept of a derivative. All quake and tremble before the ta-pocketa-pocketa of your razor wit.

      • Since Mr Bart avoids addressing Salby’s “2.6 degree” problem, he knows exactly what it is, and has not way to deal with it.

      • Wow Bart, I’m surprised at you. You would think you’d look at the flaws in Salby’s “theory”

        Here goes.

        From around 1850 until today, global temps have risen 0.7 degrees C
        From around 1850 until today, CO2 has risen 120 ppm.
        According to Salby, this temperature increase is the cause of the rising CO2.

        OK…..lets assume he’s correct.
        Today CO2 is at 400 ppm.
        If global temperatures drop 0.7 degrees C, CO2 should go down to 280 ppm
        If global temperatures drop another 0.7 degrees C, CO2 will then drop to……160 ppm
        If global temperatures drop an additional 0.7 degrees C, CO2 will then be at……40 ppm

        We’ve now dropped temps 2.1 degrees.
        Guess what happens when it goes down 2.6 degrees.
        ..
        Accordingly, the Reductio ad Absurdum logic technique has to be applied in this case.

      • No, David. You are the clueless one. That is why your inputs are so ridiculous and irrelevant. This is why you are not even wrong. You aren’t even arguing the question. You are off in some la-la land of your own.

      • I cannot help you, David. You insist on being ignorant, and embarrassing yourself. There evidently is nothing I can do.

      • Bart, I suggest you cease calling me ignorant. You really need to examine the criticism of Salby’s work. I’m just pointing out to you the error he has made, and why his “theory” is laughed at. There is ample evidence that the earth has endured drops of 4 or 5 degrees in the recent past, yet according to Salby, all life on earth would die off if temperatures dropped 2.5 or 2.6 degrees, when the concentration of CO2 would hit zero.
        ..
        But, you are free to ignore criticism if you choose. That is your choice. You are doing much the same as Salby does when he dismisses evidence that contradicts his “theory’ You can continually post the bogus WFT graph that shows the correlation between temperature and the carrying CO2 concentrations along the long term trend after it has been removed using the “derivative” function. …..Try and keep in mind that this “correlation” you post is just that, it doesn’t show causation.

      • Bart, as usual, the human emissions are ~9 GtC/year, the current pressure difference (110 ppmv) dependent sink rate is ~4.5 ppmv/year, or a e-fold decay rate of ~50 years. That is not fast enough to remove the human input. Or any excess natural input.
        Temperature is only responsible for the small variations in the rate of change which is from different processes than the longer term increase. There is not a shred of evidence that the natural inputs increased over the past 55 years, there is only evidence that human emissions increased a 4-fold over that time frame, so did the increase in the atmosphere and the sink rate.

      • You’re not offering criticism, David. You’re offering gibberish. You have a fundamental misunderstanding of the argument, and no apparent inclination to rectify your misapprehensions. You’re basically just trolling.

      • As usual, Ferdinand, you offer up only assertions.
        You’ve got to match the phase, Ferdinand. If you cannot do that, then your model is wrong.
        Temperatures match both the long term trend and the variations, with the proper phasing, in the rate of change of CO2. Human inputs have little effect.

      • Bart, I have provided you with the argument used to obliterate Salby’s entire work.

        It is not “gibberish”…it’s exactly the argument most competent scientists use to criticize Salby’s work.

      • It’s gibberish, David. And, any scientist employing your argument is incompetent. Now, will you quit trolling?

      • Bart, it is not “gibberish”
        ..
        In the words of te John Nielsen-Gammon…
        ..
        “Eventually I realized that if 0.8 C of warming is sufficient to produce an increase of 120ppm CO2, as Salby asserted, then the converse would also have to be true. During the last glacial maximum, when global temperatures were indisputably several degrees cooler than today, the atmospheric CO2 concentration must have been negative. ”
        Reference: http://atmo.tamu.edu/profile/JNielsen-Gammon

      • That’s a stupid argument. Firstly, it is fundamentally wrong because it ascribes a specific level of CO2 to a given temperature change. The theory is not that CO2 change is proportional to temperature change, it is that the rate of change of CO2 is proportional to temperature anomaly (actually, affinely related, unless you baseline the temperature anomaly to the equilibrium level).
        Moreover, if is extrapolating a local function well beyond the confines of its established region of validity. This is like saying a transistor cannot possibly work because the current shuts down when you apply a negative bias voltage across the junction.
        Really, really stupid.

      • Bart….
        “it is that the rate of change of CO2 is proportional to temperature anomaly”

        120 ppm / 150 years is the “rate of change of CO2”
        ,,
        roughly calculated like this: ( 400 ppm – 280 ppm) / (2014 – 1859)
        0.7 degrees is the temperature anomaly change in the past 150 years.
        So according to you, the 120 ppm is proportional to 0.7…….

        Now, go re-read my argument.

      • Bart, you can call John Nielsen-Gammon’s argument “stupid”….but he has a lot more credibility than you have.

      • Bart
        John Nielsen-Gammon has a lot more credibility than you have when it comes to these matters.

      • Reality must be problematic for you Bart.

        Salby’s theory is garbage. It leads to illogical conclusions. You didn’t even know about the “2.6 degree problem” it until I told you. You need to examine all sides of an purported “theory” and it’s ramifications before you latch on to it. Especially a theory that isn’t supported by real world evidence.

      • Only in your imagination, David. There is no “2.6 deg” problem. The only problem is with your understanding.

      • My understanding is very clear. The response Gammon gives is pretty simple to understand. Salby’s “theory” has a big big problem with attributing the increase of CO2 to temperature.
        …..
        This is one problem. There are others, which Ferdinand is doing a great job of addressing If you can’t address the issue more than 2.6 degrees, I can understand, it is difficult to face the fact that the theory is illogical. But it is enjoyable watching you cling to an illogical theory.

      • Your understanding is very clearly wrong. So is Gammon’s. Ferdinand is just weaving a narrative. It fails the empirical test.

      • It may make more “sense”, i.e., is easier for you to grasp. But, simple answers are often wrong answers in this complex world, and you do not have real world experience with which to gauge which hypotheses are actually more or less likely.
        The empirical observation is that, to a very high degree of fidelity, in the modern era since 1958, atmospheric CO2 concentration is the solution of the differential equation
        dCO2/dt = k*(T – Teq)
        where k is a coupling factor, and Teq is an equilibrium temperature at which CO2 would hypothetically cease increasing.
        This is a first order, affine expansion of whatever actual dynamics hold. The values of k and Teq are valid within a to be determined neighborhood of the current operating condition, but since at least 1958, they are well represented by specific, constant values depending up which temperature set is used to evaluate them.
        For the UAH lower tropospheric mean temperature record, the best values are k = 0.22 ppmv/K/month, and Teq = -0.64 degC, and the agreement over the time interval is very good.
        It does not invalidate the model that different coefficients are needed for different temperature sets, any more than it invalidates the temperature sets themselves.
        It does not invalidate the model that the same coefficients that hold in the modern era might provide questionable results over geologic time. It is a linearized model, which only holds within the neighborhood of a particular operating point. It is standard engineering practice to use such local parametrizations to model complex phenomena.
        It tells us what has been happening in the modern era since at least 1958. That is all we need to see that it accounts for essentially all the CO2 dynamics in that era. Given a record of temperatures over that time and a starting condition, CO2 levels can be estimated to a high degree of fidelity at any time within the interval – human inputs are superfluous.
        In reality, the differential equation above describes the evolution of the “equilibrium”, or attractor, level of CO2. A more complete system model would be something along the lines of
        dCO2/dt = (CO2eq – CO2)/tau + H
        dCO2eq/dt = k*(T – Teq)
        H represents human inputs, and tau represents the dominant time constant under which inputs are sequestered in the natural sinks. For tau short, the contribution of H to overall CO2 approaches H*tau, which is small, and CO2 tracks CO2. This is a very ordinary and commonly encountered feedback dynamic.

      • Your “model” is descriptive only….even you admit it when you post “The empirical observation is that, to a very high degree of fidelity, in the modern era since 1958”

        You have no empirical observations dealing with a) decreasing temperatures and/or b) decreasing CO2. Can’t test a “model” without empirical data for those specific circumstances.
        ..
        Furthermore your “evidence” uses removes the short term variation when you apply the 12-month mean to the CO2 data, and it removes the long term trend when you use the “derivative”……what does that leave you with? Your “evidence” is a correlation, and is does not prove causation.


        Secondly you post “It does not invalidate the model that the same coefficients that hold in the modern era might provide questionable results over geologic time” …..so in effect all you are doing is curve fitting recent data to your model”
        ..
        Your model is useless if it is “sensitive” to the time interval (i.e. non-geologic) applied. Basically it offers no predictive capability.
        Notwithstanding the root of the problem is that Salby’s theory is still garbage.

      • “You have no empirical observations dealing with a) decreasing temperatures and/or b) decreasing CO2.”
        You would disqualify everyone’s models by that standard.
        Look, the correlation holds over a wide range of inputs which bobble up and down, giving a full frequency band of observations. The relationship holds. It tells us humans are not driving atmospheric levels.
        “Furthermore your “evidence” uses removes the short term variation when you apply the 12-month mean to the CO2 data…”
        It removes, specifically, the annual cycle, which is much more variable in the rate of change than any other component, and swamps out any long term relationship. To get the long term relationship, you have to filter it out. There is nothing unusual or underhanded about that. It’s standard practice.
        “…and it removes the long term trend when you use the “derivative”…”
        You keep harping on that, and seem not to be able to understand that it is immaterial. The equation
        dCO2/dt = k*(T – Teq)
        matches the data, including the long term trend in dCO2/dt. It is that long term trend in dCO2/dt which invalidates human attribution. It is accounted for by the temperature relationship. Human inputs also have a long term trend in the rate of emissions. There is little to no room for it. It is already accounted for. Hence, human inputs are not a significant driver.
        When you integrate the equation, you get everything back. This criticism holds no water.
        “…so in effect all you are doing is curve fitting recent data to your model.”
        I am validating a local model by fitting it to the data. All models have to validate themselves against the data. There is nothing wrong with that. The important part is, it is a complex fit – lots of bumps and burbles along the path, and they all match. Compared to a superficial match of human inputs with concentration, which only seem to vaguely match when you take out all the bumps and burbles, it’s a slam dunk which model is right.
        “Basically it offers no predictive capability. “
        It offers predictive capability within a local timeframe. It said that the rate of CO2 would decline when temperatures stalled, and they did. It says the rate of change will decline with declining temperatures. Watch and see what happens.

      • 1) Yup…It does and it invalidates yours too.
        .
        2) “the correlation holds” , but it is only a correlation. Doesn’t tell us anything about what is driving atmospheric CO2. All it shows it that the “wiggle” of variation from the upward linear trend is correlated to temperature. It says nothing about the upward linear trend.
        .
        3) Removing the annual cycle in the CO2 ? But you don’t do the same for temperature……you are now comparing apples too orange.
        4) “To get the long term relationship, you have to filter it out.” Nope… the long term trend is clearly visible in the data without removing the annual variation With my eyeballs, I see a rise of 80 ppm over 50 years.
        http://www.woodfortrees.org/plot/esrl-co2
        ( for future reference use the “mean” on UAH data also —-> http://www.woodfortrees.org/plot/uah/plot/uah/mean:12 )
        5) ” that it is immaterial” .. no, sir, that is THE POINT. Your “model” does not capture the relationship between temperature and the long term trend. ( which is why your model is useless for predictive capability)
        6) “This criticism holds no water.” When you integrate temperature over time, you get a physical meaningless quantity
        ..
        7) You are “curve fitting”….in fact you are doing nothing more than y=mx+b to find the different coefficients for the different datasets (scale and offset) . The fact that you need different ones for RSS, GISS, etc proves you are not validating anything. It proves your “model” is ……uh…….flaky
        8) All of the “bumps and burbles” are correlated to temp. It’s not causation, and the long term upward trend of CO2 is missing
        ..
        9) “It offers predictive capability within a local time frame.” …..it sure didn’t predict the 30 ppm rise in CO2 in the past 15 years with the “pause” in the rise of global temps. Which by the way, makes your model……worthless.
        10) “It said that the rate of CO2 would decline when temperatures stalled, and they did.” …….Mother Nature disagrees with you on that point….. (looks like it accelerated)
        http://www.woodfortrees.org/plot/esrl-co2/from:1990/plot/esrl-co2/from:1990/trend/plot/esrl-co2/from:2000/trend

      • PS….”physically meaningless ” is not the same as “useful”

        Degree-days are used extensively in the heating/cooling industries, but that is relative to an assumed heat gain (loss) that varies with temperate.
        The example that shows it is “physically meaningless” is as follows
        ..
        Take two Dewar flasks and fill one with 30 degree C water and the other with 50 degree C water. After three hours remove the 50 degree water from it’s flask. after another two hours (for a total of 5) remove the water from the 30 degree flask. Integration of the values in these two distinct situations say that both had 150 degree-hours.

      • 1) Technological society depends on modeling techniques such as I am using. You can deny the successful application of these techniques if you like, but it’s pretty much a non-starter.
        2) Sure it does. It tells us a temperature modulated process is the main driver. Now, you hunt for such a process and start gathering data and doing experiments. There are many potentialities, deep ocean upwelling being a rather prominent one.
        3) Actually, the same is done for temperatures, because these are temperature anomalies. They have yearly variation taken out already.
        4) The trend in total CO2 tells you nothing about its source.
        5) It captures the long term relationship between temperature and the long term trend in the rate of change of CO2. If you want the long term trend in overall CO2, you simply integrate that relationship.
        6) No you don’t. The coupling constant has the proper units to produce ppmv. This is the same as any process. E.g., the amount you press down on your gas pedal translates into a speed on the highway. If you integrate the inches you depress it, you get inch-hours, or some such “meaningless” quantity. So, you multiply it by the scale factor which tells you how many mph you get per inch, and voila, you have miles traveled.
        7) Nonsense. You could as easily say that the fact that UAH and GISS don’t agree says they are both invalid.
        8) The bumps and burbles in temperature precede the bumps in burbles in CO2 (i.e., they are coincident with the bumps and burbles in dCO2/dt). That points the arrow of causation in the direction of temperature to CO2. There are beaucoup outgassing processes whose rates are determined by temperatures – it’s is a very reasonable inference. And, finally, the upward “trend” can be retrieved through integration. It’s still there, it’s just become a constant offset. If you knew calculus, you would know this is a trivial matter.
        9) Yes, it did. Look at the plot.
        10) Read more carefully. Look at the plot. The RATE of CO2 has stalled. Human emissions haven’t. Human emissions are not driving this.
        http://i1136.photobucket.com/albums/n488/Bartemis/CO2_zps330ee8fa.jpg

      • David Socrates @ January 28, 2015 at 4:24 pm
        The example that shows it is “physically meaningless” is as follows
        An example, and a completely irrelevant one.
        Turn up your oven to 400 deg. Put a pot of low temperature water in it for a minute. Take it out and measure the temperature. Now, put another pot in for two minutes and take it out. Repeat for any time period for which the water stays well below 400 deg (i.e., before the water has had time to approach steady state). You will find that the temperature change upon removal is proportional to the oven temperature times the time it stayed in there.
        This is a system which obeys a differential relationship of the form
        dT/dt = (400 – T)/tau
        where tau is the thermal time constant. Prior to steady state, when T is much less than 400, it is approximately
        dT/dt := 400/tau
        Integrate the oven temperature in degree-minutes (400 times the length of time in this case where the oven temperature is constant) and divide by the time constant, and you will get the approximate temperature change of the water in the pot upon removal.
        This is really basic stuff.

      • 1) Stop moving the goal posts. Validation is the subject here, most of the applied models are validated before application.
        .
        2) It’s a correlation, is says nothing about causation (PS also there’s a boatload of down dwelling at the poles too)
        .
        3) Salby’s theory and your model goes in the toilet when absolute temperatures are used.
        4) The trend in total CO2 tells you nothing about its source???? It says the sources are overwhelming the sinks. Funny how human emissions are twice the rate of atmospheric accumulation. That says nature is not sinking all of human emissions.
        ..
        5) The integral of temperature over time has no meaning
        ..
        6) “proper units” does not prove anything. It means you didn’t make an error when you picked them.
        .
        7) You are using UAH, RSS, etc as evidence Changing your “model” to fit the evidence is …..known as “curve fitting”
        ..
        8) “The bumps and burbles in temperature precede the bumps in burbles ” Not always….sometimes it lags, sometimes its coincident, and sometime it leads. That’s why your “model” is merely correlational , and does not show causation.
        ..
        9) The integration of temperature over time is meaningless.
        ..
        10) The wiggly measured ppm line has too much variance in the data for you to make that claim. Try using trend lines for both sets of data.

      • PS… in your graph of the rate of atm CO2 vs Human Emissions…
        ..
        Apples and oranges.

        Try comparing the rate of change of atmospheric CO2 with the rate of change of human emissions.

      • 1) no merit
        2) no merit
        3) There is no “absolute temperature” for the Earth. Comment has no merit.
        4) Discredited pseudo-mass balance argument. No merit.
        5) I showed how it does.
        6) stupid
        7) stupider
        8) Obviously, writer has no experience dealing with noisy, real world data
        9) stupidest
        10) The flatlining of the rate of change of CO2 is coincident with the stall in temperatures. It’s real.
        “Try comparing the rate of change of atmospheric CO2 with the rate of change of human emissions.”
        It is the rate of change of human emissions. Look at the labels, for crying out loud.
        This is going nowhere. You are intent on remaining clueless. Have at it.

      • “socrates” says:
        John Nielsen-Gammon… has a lot more credibility than you have.
        John N-G agrees with Prof. Robert G. Brown, a regular contributor here. He quotes Brown that skeptics…
        …do not ‘deny’ AGW…What they challenge is the catastrophic label and the alleged magnitude of the projected warming on a doubling of CO2.
        So far, the real world agrees with skeptics of man-made global warming [MMGW]. That is the big difference between climate alarmists like socrates, and rational skeptics who look at the evidence — skeptics will change their minds if new evidence tells them to.
        Alarmists like ‘socrates’, OTOH, will never change their minds. They can’t. Once an alarmist buys into the Narrative, they are psychologically incapable of changing their minds. That’s why Bart says: This is going nowhere. Alarmists are religious converts. As such, they cannot be scientists.
        John N-G points out that for many millions of years…
        …the Antarctic was ice-free. The climate state associated with the ice-free Antarctic was warmer by several degrees
        Where was the high CO2? In fact, CO2 was low during those times. Any scientific skeptic would look at that and conclude that the ‘CO2=MMGW’ argument has been solidly deconstructed. [For the record, I think that CO2 has a *very* minuscule effect on global T. CO2 is only a very small 3rd-order forcing, which can be completely disregarded in the context of the current ‘carbon’ scare.]
        John N-G has a 6-part series that begins here.
        WUWT reported on John N-G here and in other articles. He is much closer to being a scientific skeptic than even being a lukewarm alarmist. So anyone commenting on him is probably not very familiar with where he stands…

      • 1) I win
        2) I win again
        3) https://wattsupwiththat.files.wordpress.com/2013/03/clip_image0041.jpg
        4) I love this one
        Let CO2 = annual increase of atmospheric CO2
        Let H = Annual human emission of CO2
        Let N = Annual natural emission of CO2
        Let U = Annual natural uptake of CO2

        Then CO2= H + N – U
        and CO2 – H = N – U
        ..
        Since H = 2*CO2
        Then CO2- 2*CO2 = N – U
        or – CO2 = N – U

        Now what does this say?
        It says that because CO2 has risen in the atmosphere by about 2 ppmv per year, – CO2 is a negative number
        ..
        and that means N – U is negative, which says nature is absorbing more CO2 than it is releasing. Not only is it doing that, it is doing that while temperature is rising. QED
        5) http://wattsupwiththat.com/2015/01/25/an-engineers-ice-core-thought-experiment/#comment-1846464

        6) I win
        ..
        7) I win
        ..
        8) Obviously you have no experience with the real world
        ..
        9) I just keep on winning
        ..
        10) Y-axis is labeled “Global emissions Mt/yr” That is not a rate of change that is a measured quantity. The rate of change of human emissions would be Mt/yr^2 For example if humans emitted 4000 in 1970 and 4300 in 1971, the rate of change would be 300 Mt/yr PER YEAR.

      • “Where was the high CO2? In fact, CO2 was low during those times”

        Citation?

        PS while you are at it, how about the 20X citation……could that have been the cause of an ice free Antarctica?

      • “and that means N – U is negative, which says nature is absorbing more CO2 than it is releasing.”
        Utterly inane.
        N = Na + Nn
        Na = response of natural sinks to anthropogenic forcing
        Nn = response of natural sinks to natural forcing
        Na + Nn – U is negative… What does it mean? Nothing. You have two variables and only one equation, and cannot solve for both of them.
        Anthropogenic attribution is only valid if Nn – U is negative.
        This is a dynamic system. Only idiots who have no understanding of dynamic systems fall for the pseudo-mass balance argument.
        ” The rate of change of human emissions would be Mt/yr^2…”
        Too stupid for words.

      • Stupid nomenclature as well.
        “and that means N – U is negative, which says nature is absorbing more CO2 than it is releasing.”
        Utterly inane.
        U = Ua + Un
        Ua = response of natural sinks to anthropogenic forcing
        Un = response of natural sinks to natural forcing
        N – (Ua + Un) is negative… What does it mean? Nothing. You have two variables and only one equation, and cannot solve for both of them.
        Anthropogenic attribution is only valid if N – Un is negative.
        This is a dynamic system. Only idiots who have no understanding of dynamic systems falls for the pseudo-mass balance argument.

      • @”socrates”:
        I win everything! Doubled and squared!
        There’s my assertion, just like yours. And just as (in)credible.
        soxie sez:
        PS while you are at it, how about the 20X citation
        Citation sitting right here and waiting for your agreement that you were wrong to dispute it. Just post that you will admit your error if I post the citation, and you will get the citation. Not before.
        Re: the ’cause of an ice free Arctic’, is on the wrong time scale. But nice try, and thanx for playing!

      • Bart…

        N is natural emissions. Measured in Gt. There is no need to consider splitting it into two classes.
        U is natural uptake measured in Gt….Again, there is no reason, nor is there a need to split it.

        You are making something that is inherently simple using plain old arithmetic
        ..
        When nature absorbs CO2, it doesn’t care where it came from.
        And I already have seperated out the source as being either natural or of human origin.
        ..
        Yup, it is dynamic. N and U are very dynamic. But the beauty of the mass balance argument is this.
        ..
        We can measure CO2, and we can measure H. That is why we know the left hand term CO2 – H is negative.

        PS….I will also comment on the fact that you can’t read.
        ..
        You posted

        Na = response of natural sinks to anthropogenic forcing
        Nn = response of natural sinks to natural forcing


        and you posted

        Ua = response of natural sinks to anthropogenic forcing
        Un = response of natural sinks to natural forcing


        So, according to you Na=Ua and Nn=Un


        See?……in your world everything is a “sink”

      • Dbstealey….
        ..
        Could you please stalk me some other time? I’m trying to have a conversation with Bart.

      • soxie,
        You should give up when you’re only that far behind. Because Bart is killing you.
        [As for ‘stalking’, heh, I’ve pointed out that you are fixated on me at least a dozen times now. Your last comment is just the latest example of your fixation. Look up psychological “projection, and Example #1 will be ‘socrates’.]
        And any time you want, I have the CO2 link you’re so desperate to have me post. The usual condition applies.
        So now that you’ve painted yourself into a corner with your sad projection, here’s the deal: don’t comment on my posts, and I won’t respond to yours. I am more than happy to have you stop endlessly bird-dogging my posts with your insecure pseudo-science. So I trust this will be the last post between us.

      • Stop being a churl as well as a troll, David. The first response was written in haste. Go by the second one.
        U has two components, Ua and Un. Ua is wholly due to H. Ua is created as a response to H. If H ceases, then Ua ceases, too. That is the nature of a dynamic system.
        Only Un is wholly natural. Only the sign of N – Un can resolve anthropogenic attribution. There is not enough information here to resolve it.
        That is the mistake the stupid people who promote the pseudo-mass balance argument make. They are idiots. If you agree with them, then res ipsa loquitur.

      • CO2 = N + H – Ua – Un

        Right?
        CO2 – H = N – Ua – Un

        Right?

        We know CO2 – H is negative.
        That makes N – Ua – Un negative.

        What does that mean?
        ..
        It means that natural emission minus ALL uptake processes are negative.
        So in spite of rising temperatures, nature is uptaking more CO2 than it is releasing.

      • Which is meaningless. Only if N – Un is negative can you say that CO2 could not have expanded without humans. Once again, Ua is a response to H. It exists wholly because of H. In a very real sense, it is artificial sink activity, because it is induced by human activity.
        Do you know that N – Un is negative? No. You only know that N – Ua – Un is negative. That’s not enough to assign attribution.
        If sinks are very reactive (and, all indications are that they are), then it can be as high as H itself, and everything you see would be exactly the same with or without humans.

      • If sinks are very reactive (and, all indications are that they are), then Ua can be as high as H itself, and everything you see would be exactly the same with or without humans.
        dbstealey @ January 28, 2015 at 8:06 pm
        He’s just fumbling around, and trying to play rhetorical games. He doesn’t really understand anything. He just wants to annoy.

      • Bart…

        “Only if N – Un is negative can you say that CO2 could not have expanded without humans”

        Strawman, I am not discussing the case of there not being any humans. Keep this discussion to an earth with humans on it.
        CO2 – H = N – Ua – Un
        The left hand side of the equation has all of the measurable quantities, and the right hand side has everything else.
        .
        .
        N – Ua – Un < 0 That is what physical measurements say.
        That says more CO2 is being taken up than is being emitted.
        There are only two things at play on the right hand side of the equation …..here, sinks and sources. You don't need to measure Ua, Un or N …..all you need to know is that the sum of all of them is negative.
        .

      • No, David. If you want to determine if humans are responsible for the observed rise, you have to establish that the rise would not have occurred without humans.
        If there were no human inputs, if there were no H, then there would be no Ua. The sinks expand in response to increases in the inputs from any source. Ua is a direct response to the H input, just as Un is a response to N.
        If the sinks are very active, then you can have Ua take out very nearly all of H. In that case, the balance is
        CO2 := N – Un
        regardless of human activity, and everything you see would have happened with or without humans. Un may take out an equal proportion of N, but N is at least 30 times larger than H, so it is fully capable of driving what has been observed.
        That is the nature of a dynamic system. That is reality. Only fools fall for the pseudo-mass balance argument.

      • Bart…..I’m not attributing the rise to humans….you’d best review all of my posts, nowhere did I make that claim. The only claim I am making is that N – Ua – Un is negative. That is what the physical measurements are saying. Those measurements say that all uptake processes exceed the natural emissions. This is happening in spite of rising temperatures. It clearly falsifies Salby.
        All your arm waving about human activity doesn’t mean anything, as it is all taken care of in the equation.

      • It says nothing of the sort, David. The only way Salby would be wrong would be if you could establish anthropogenic attribution. Since you cannot, you have nothing.

      • It says nature is taking in more CO2 than it is emitting as temperature is rising.
        ..
        It falsifies Salby, because according to him nature should be emitting more CO2 than it takes up as temperature rises.
        ..
        N – Ua – Un < 0
        ..
        That's what the measurements say.
        That's what is happening.

      • The sinks expand in response to increases in the inputs from any source. Ua is a direct response to the H input, just as Un is a response to N.
        This is a dynamic system. You cannot use static accounting for a dynamic system.
        It is really quite simple, though the uninitiated seem to have a very hard time with comprehending the concept of dynamic flows.
        No H, no Ua.
        You cannot book sink activity as though human forcing were not there. It is, and it induces additional sink activity which would not exist were it taken away.

      • Bart.
        ..
        Two things.

        H exists. Try as hard as you want, you cannot make it go away. It’s here, and it is a measurable quantity. It’s on the left hand side of the equation because it is measurable.
        ..
        Second.
        ..
        According to you ” Ua = response of natural sinks to anthropogenic forcing”

        You see? even you said the sink was natural. Can’t measure it, so it’s on the right hand side of the equation. Don’t care to measure it, don’t need to. The sum of the terms on the right is negative.
        ..

      • No H, no Ua.
        It is part of natural sinks, but it is anthropogenically induced. Ua cannot be counted in the ledger against Salby.

      • Bart
        ..
        We would not be having this discussion if H = 0

        Even if H = 0 and Ua = 0 that does not invalidate the CO2 – H = N – Ua – Un equation.
        ..
        If H were zero, there would be no measurements of CO2.

      • “Even if H = 0 and Ua = 0 that does not invalidate the CO2 – H = N – Ua – Un equation.”
        I think you’re getting sleepy. That would mean CO2 = N – Un, which would be positive, i.e., nature would be a net source.
        This isn’t that hard. Ua is not a spontaneous product of nature. It only exists because of human forcing. No H, no Ua, and you do not know what sign N – Un would be. You do not know what U is when you take out Ua.
        I do. N – Un is positive. Natural sources are currently overwhelming natural sinks. That is what the dCO2/dt-temperature data tell us. It is very clear. There is no doubt about it.
        That is my final word for now. Good night.

      • No Bart, with CO2 = N – Un, you have no way to tell if CO2 were positive or negative. With H = 0 there are no human beings. There would be nobody to measure it. There would be nobody to care.
        ..
        The reality is the H > 0
        The reality is that humans have measured CO2
        ..
        We know CO2 – H is negative. We’ve measured it.
        ..
        Now, you can build stupendous differential equations, derivatives, limits and integrals on the right hand side of the equation, but it doesn’t matter. All the measured values are on the left, all the unknowns on the right. The sum of all the terms on the right is a single value, and it has been measured to be negative.
        “N – Un is positive.” Don’t care because if it is, the Ua > N – Un It has to be, because that is what the measurements say,.

      • Not quite gone yet.
        “Don’t care because if it is, the Ua > N – Un It has to be, because that is what the measurements say.”
        Fine, but Ua is positive, so N – Un can be positive. That makes nature a net source.
        You cannot get around it, David. If you do not know the individual responses Ua and Un, then you do not know if nature is a net sink or source. You do not know if CO2 would have risen or fallen without humans.
        If, as I claim, Ua is approximately equal to H, then N-Un can be the whole enchilada. It can account for essentially the entire set of observations, with H playing an insignificant role.
        This is, in fact, the case.

      • “Ua is positive” ……don’t care
        “N – Un is positive” ….. don’t care
        “as I claim, Ua is approximately equal to H” ……don’t care
        ” N-Un can be the whole enchilada.” ….. don’t care
        ..
        All that matters is….
        Ua > N – Un that is what is measured.
        ..
        That makes N – Un – Ua negative…..
        ..
        You can’t get around the measured values that Mother Nature has provided you with.

      • ” Ua is positive” ….. don’t care
        “N – Un can be positive.”……don’t care
        “That makes nature a net source.” ….. Nope, N – Ua – Un is negative, because Ua > N – Un as measured.
        “If you do not know the individual responses Ua and Un, then you do not know if nature is a net sink or source”…..don’t care what the values of N, Un or Ua are, I do know that their sum is negative
        “You do not know if CO2 would have risen or fallen without humans.” ……again, H > 0 so you are talking about something that doesn’t matter today.
        “Ua is approximately equal to H” ……don’t care, the sum of Ua, Un and N is still negative irrespective of the value of Ua.
        ” N-Un can be the whole enchilada”……again, don’t care because you are trying to make statements about the individual values of Un, Ua, and N, and none of them can be measured. You can hypothesize about them all you want, but you have no data to back up any of your assertions.

      • ” Ua is positive” ….. don’t care
        .
        “N – Un can be positive.”……don’t care
        .
        “That makes nature a net source.” ….. Nope, N – Ua – Un is negative, because Ua > N – Un as measured.
        .
        “If you do not know the individual responses Ua and Un, then you do not know if nature is a net sink or source”…..don’t care what the values of N, Un or Ua are, I do know that their sum is negative
        .
        “You do not know if CO2 would have risen or fallen without humans.” ……again, H > 0 so you are talking about something that doesn’t matter today.
        .
        “Ua is approximately equal to H” ……don’t care, the sum of Ua, Un and N is still negative irrespective of the value of Ua.
        .
        ” N-Un can be the whole enchilada”……again, don’t care because you are trying to make statements about the individual values of Un, Ua, and N, and none of them can be measured. You can hypothesize about them all you want, but you have no data to back up any of your assertions.

      • You are quite hopeless and completely illogical. But, for any onlookers, one last nail in your casket.
        Look at it this way. Let Un = f*N and Ua = f*H for some feedback factor f between zero and one which is identical for both inputs to be treated equally, Then we have
        CO2 = (1-f)*(N + A)
        Observed CO2 rise is about 0.5*A, so the lowest f to be consistent with observations is 0.5. This is the weak sink case, in which all of the observed rise is anthropogenic.
        As f increases above 0.5, more and more of the observed rise has to have a significant natural component. If f = 1 – eps, for some small value eps (powerful sink activity), then
        CO2 = eps*(N+A)
        and the rise is necessarily almost wholly from N. Since we know N is some 30 times larger than A, we could have eps as small as 1/60, in which case only 1/60th of A gets into the observed rise.
        It all depends on the power of the sinks. And that is why the pseudo-mass balance argument is a failure. It is essentially an assumption of weak sink activity, and a train of circular logic from there.
        The dCO2/dt-temperature relationship tells us that sink activity is powerful, and human inputs have little impact.

      • Bart…
        ..
        You can’t measure f
        ..
        You failed to define what “A” is
        ..
        Please re-write your post so that it makes sense

      • CO2 = (1-f)*(N + H)
        CO2 = N + H -fN -fH
        CO2 – H = N -fN – fH ( which is same as CO2 – H = N – Ua – Un)
        “so the lowest f to be consistent with observations is 0.5” The value of “f” doesn’t change the equation, and you don’t need to know anything more than the sum is still negative
        The basic fact measured is that CO2 – H is negative
        ..
        You can play all you want with extra values and such on the right hand side of the equation, but it doesn’t matter. Whatever value you end up with in your musings it is negative due to measured values
        “Since we know N is some 30 times larger than A,”…..doesn’t matter if it is 29.3 or 31.6…….the right hand side of the equation is still negative
        All the dCO2/dt-temperature relationship tells us is that there is a correlation between temperature than the variance around a rising level of CO2

      • Bart:
        CO2 = (1-f)*(N + A)
        That is only true if the CO2 is directly proportional to the momentary inputs, thus with extreme fast response of the sinks. That is only true for the ocean surface and seasonal + short term for vegetation as response to temperature. For the ocean surface also for increased CO2 pressure in the atmosphere, but at 10% of the change in the atmosphere the ocean surface is saturated. 90% of the sink rate is caused by much slower processes.
        As there is an increase in the atmosphere noticed, the increase responds to the total increase in the atmosphere, not to the momentary inputs with a much slower response time. That means that all increase is caused by A, as ΣN is negative:
        CO2 = (1-f)*((ΣN+ΣA)-Co)
        where Co is the CO2 level at equilibrium and ΣN+ΣA the total net increase of CO2 in the atmosphere.
        Except if N increased at exactly the same timing and with the same ratio as A increased over time (a 4-fold since 1959) for which is not the slightest evidence…
        Your arbitrary splitting of the sinks in “human caused” and “natural caused” has no effect: if humans emit 5% of total emissions, they are responsible for 5% of the sinks with your fast response, which means that the carbon balance of “natural” still is negative:
        Total sinks = 0.5 * A
        Human part = 0.025 * A
        Natural part = 0.475 * A
        Still more natural sinks than sources and zero contribution of N to the increase in the atmosphere…

      • David doesn’t get the argument. Again. Who’da seen that coming?
        Ferdinand:
        “That is only true if the CO2 is directly proportional to the momentary inputs, thus with extreme fast response of the sinks.”
        As I explained, f encompasses both fast and slow response of the sinks. If they were slow, most of the rise would be anthropogenic. When they are fast, most of the rise is natural.
        This is a general feedback scenario, which is widely applicable. In the case where the feedback is more involved than a simple proportionality factor, it can be considered the dc gain of an operator.
        “…as ΣN is negative…”
        Circular logic with no justification.
        “…for which is not the slightest evidence…”
        Nonsense. You keep saying that, and I keep offering the evidence. Just because you want it to be does not make it so.
        “…if humans emit 5% of total emissions, they are responsible for 5% of the sinks with your fast response…”
        Sorry, no. I explained this. If sinks are fast, the rise is almost all natural. There is no valid argument against it, and you have not offered, indeed cannot offer, any. It is not open to negotiation. It is math.

      • Bart,
        All you have is a pure theoretical idea, based on curve fitting, which fails about all known observations…
        1. The theory is based on one process that fits all: short term variations and long term trends.
        That is demonstrably false: the short term variability is from the short influence of temperature on vegetation, the long term trend is not from the influence of temperature on vegetation (which is opposite).
        2. The theory is based on a huge change in ocean CO2 releases.
        That is demonstrably false: there is no decrease in residence time of CO2 in the atmosphere and no increase of δ13C if the ocean input/throughput was increasing.
        3. The theory says: dCO2/dt = k*(T – Teq)
        That is demonstrably false: Your formula means that any small sustained change of temperature will give a continuous input of CO2 in the atmosphere without any feedback from the increased CO2 pressure in the atmosphere. Any disturbance of the CO2 input, be it natural or human is met with a reaction from the sinks from the changes in CO2 pressure in the atmosphere: any increase in temperature or ocean input will give a finite increase in the atmosphere at a higher CO2 level, not an infinite increase.
        4. The theory says that the natural increase in CO2 dwarfs the human input.
        That is demonstrably false: The estimates for the current throughput are:
        ~40 GtC continuous in from ocean upwelling and out at ocean sink places
        ~50 GtC in/out seasonal for the ocean surface
        ~60 GtC out/in for vegetation
        ~10 GtC in from humans
        Or human emissions are one-way ~6.5% of the total carbon cycle, small but detectable (and measured).
        Human emissions increased a 4-fold 1959-2012, so did the increase in the atmosphere, that means that the net sinks also increased a 4-fold in that period.
        To increase the sinks a 4-fold, if that was caused by a natural influx, that should have increased a 4-fold too in the same period, as the sinks don’t make a differentiation between human and natural CO2. But that is not what is observed in 1. and 2.
        5. The theory says: If sinks are fast, the rise is almost all natural.
        That is demonstrably false: sinks mainly depend of the increased pressure in the atmosphere. The current increase is ~110 ppmv above the temperature equilibrium for the current temperature, the current net sink rate is ~2.15 ppmv/year, that gives an e-fold decay rate for the excess CO2 of ~51 years.
        6. Whatever you do by splitting the sinks in “natural” and “human”, the natural outputs are always larger than the natural inputs, thus all the increase is entirely from the human emissions.
        As said by some more learned than I am: any beautiful theory can be falsified by one ugly fact.
        In this case all facts are against your theory and not one fact supports it, except for a nice fit of noise around a trend, which doesn’t give you any clue of what caused the trend itself…

      • 1) assertion
        2) assertion
        3) Formula obviously fits. Continuous transport of fresh CO2 into the system produces a continuous rise. Were it not so, fresh CO2 from human emissions could not cause a rise, either.
        4) Everyone knows human inputs are at best about 3% of natural flows. You misunderstand the action of a feedback system. All that is needed is a relatively short tau in the expanded system
        dCO2/dt = (CO2eq – CO2)/tau + H
        dCO2eq/dt = k*(T – Teq)
        This is not in any way remarkable.
        5) I showed it previously on this page. As power of sinks increase, human inputs are taken out more and more rapidly. At some point, more than 1/2 is taken out in short order, and the rise has to depend more and more on natural forcing.
        6) Nope. Fallacious “mass balance” argument. If the sinks are very powerful, human inputs cannot be responsible for the rise.
        You’ve given few facts. Much assertion. Several instances of fallacious logic.

  49. mpainter January 27, 2015 at 2:21 am

    Oh, crap, Willis, the issue is his qualifications. Dispute his conclusions, fine I don’t give a hoot; I never commented on anything but his qualifications. So I will pass on your attempt to inveigle me in issues of science.

    Yeah, it wouldn’t do for you to actually get involved in nasty old science, those facts are stubborn things.
    As to his qualifications, you still haven’t said what they are.

    I will only note that he is not the only qualified scientist who questions the reliability of ice core CO2 data.

    What, now you are going to argue “consensus”? Really? In any case, that’s just handwaving until you give us the names of a couple of “qualified scientists” who agree with Jaworowski.

    I note that you have the decency to withdraw your comment that he was a “crackpot” and I approve.

    Thanks, I admit, it was over the top.
    mpainter January 27, 2015 at 2:35 am

    Dig deeper, Ferdinand. By his own profession, he organized about ten glacial expeditions to recover ice cores for study. He was one of the first to analyze ice deposits for pollution.

    This is by my count the third time you’ve been asked for evidence to back up your claims, and you’ve responded by waving your hands and telling us some variation of “Dig deeper”. Both Ferdinand and I have dug and found nothing. My friend, those are YOUR claims, not Ferdinand’s or mine. You inability to support them is becoming quite obvious. Where is the evidence that he organized any expeditions looking for CO2? As I said above, I found two papers relating to metals and radioactivity in ice cores, but NOTHING about any expeditions regarding CO2.

    I will stand on my original statement in this regard. Please read it again, carefully, and perhaps we can end this absurd discussion, thank you.

    I haven’t a clue what you are calling your “original statement”. If you can’t be bothered to quote others, at least quote yourself so we can see what you are referring to. If it was this one:

    I am not so quick to dismiss Jaworowski who has decades of ice core experience and expertise internationally recognized.

    then you have not provided a scrap of evidence to back it up … but you may not be referring to that statement.
    Look, I think everyone agrees that he was an expert on radioactivity. But regarding CO2 he was just wrong, is NOT an internationally recognized expert, and has made a number of claims that are totally ludicrous.
    w.

  50. Regardless of possible catastrophic cooling at the start of interglacial periods in the past, our solar system entered a spiral arm of the galaxy 15.8kya(1), encountering interstellar dust which deflect cosmic ray bombardments.
    Previously, our solar system’s entry into the local bubble, a region of space devoid of interstellar matter, caused earth to be intermittently bombarded by powerful cosmic ray blasts (2)(3).
    Constant and extensive monitoring of the gulf stream current has not detected any anomalous activity as of 2011(4).
    References:
    1) Evidence for a Global Warming at the Termination I Boundary and Its Possible Cosmic Dust Cause
    2) Introduction: Paleoheliosphere versus paleolism
    3) Solar Journey: The Significance of Our Galactic Environment for the Heliosphere and Earth
    4) Monitoring the Atlantic meridional overturning circulation

  51. Arbitrating discussions wherein “Rank before Frank” is the primary qualifier of “what is” or ”what isn’t” actual, factual, reasonable and/or logical Science …. serves no purpose whatsoever in aiding or furthering the knowledge of the natural world.

  52. Samuel C Cogar January 27, 2015 at 1:44 pm Edit

    In reference to post January 26, 2015 at 10:30 pm

    …. the addition of 2.18 Gtonnes of C changes the atmospheric concentration by 1 ppmv … if not let me know. Note that this is 2.18 Gt of C that remains in the atmosphere, not 2.18 Gt emitted.

    My simple math calculation when atmospheric CO2 was at 393 ppm …. resulted in the fact that …. 1 ppm of atmospheric CO2 is equal to 5 Gts, ….. to wit:
    The average mass of the atmosphere is about 5 quadrillion (5,000,000,000,000,000) metric tons.
    The Carbon dioxide (CO2) in the atmosphere is currently at 393 ppm — or 0.0393%
    Thus, @ 0.0393%, there is about 1,965,000,000,000 metric tons of CO2 in the atmosphere.
    Or, there is about 1,965 billion tons of CO2 in the atmosphere.
    Therefore, @ 393 ppm, each ppm of CO2 is equal to about 5 billion tons of CO2.
    But now, just the weight of the carbon (C) in the CO2 would be 1.36 Gt per 1 ppm of atmospheric CO2 …… to wit:
    CO2 = 44. … (2 O2 at 16 and 1 C at 12) … and weight of 1 C is 27% of the CO2 weight.

    The difference is that you’ve confused ppmv, parts per million by volume, with ppmw, parts per million by weight.
    5.15E+15 mass of atmosphere
    1.00E-06 = one ppmv, which converts to 1.55E-06 ppmw, times atmospheric mass gives us
    8.01E+09 tonnes CO2 / ppmv, which is
    2.18E+09 tonnes of C / ppmv

    • Willis Eschenbach January 27, 2015 at 5:28 pm

      The difference is that you’ve confused ppmv, parts per million by volume, with ppmw, parts per million by weight.
      5.15E+15 mass of atmosphere
      1.00E-06 = one ppmv, which converts to 1.55E-06 ppmw, times atmospheric mass gives us
      8.01E+09 tonnes CO2 / ppmv, which is
      2.18E+09 tonnes of C / ppmv

      Willis, if I am confused ….. then what about Ferdinand, to wit, your original statement followed by his response:

      Willis E – January 26, 2015 at 10:30 pm
      Robert, I think the number you are looking for is that the addition of 2.18 Gtonnes of C changes the atmospheric concentration by 1 ppmv … if not let me know. Note that this is 2.18 Gt of C that remains in the atmosphere, not 2.18 Gt emitted.
      http://wattsupwiththat.com/2015/01/25/an-engineers-ice-core-thought-experiment/#comment-1845010
      Ferdinand – January 27, 2015 at 1:46 am
      Cumulative emissions are the sum of all human emissions translated from GtC fossil fuel use to ppmv (factor 2.12 GtC/ppmv). ………….. (snip) …………… Thus some of the extra CO2 is removed. For the current extra CO2 (110 ppmv) above the temperature driven equilibrium, and emissions – increase in the atmosphere is about 2.15 ppmv/year that gives:

      And then I found this conversion rate which is different than both the above, to wit:

      Global Commons Institute – UK
      The data plotted again below show the rising atmospheric concentration of CO2 [1959-2011] only this time converted to ‘Gt C’ or ‘giga [billions] of tonnes of carbon’. The standard conversion rate is 2.13 Gt C is equal to one ppmv CO2
      http://www.gci.org.uk/Documents/CO2_MLO.pdf

      One thing, Willis, I can not figure out where you got your converted rate of “1.55E-06 ppmw”.
      But anyway, I am SKEPTICAL that someone is trying pull that ole “shell game” on me but I’m not quite sure as yet.
      Willis, me suspects that your above calculations result of …. 2.18E+09 tonnes of C / ppmv ….. is wrong, simply because it appears to me that you converted @ STP (sea level) and then multiplied by total atmospheric mass,
      Does not your stated “1.55E-06 ppmw” continue to decrease as you ascend into the atmosphere, just like the “air pressure” does?
      To wit, for further reading:
      ppmv – David Randall
      http://kiwi.atmos.colostate.edu/group/dave/pdf/ppmv.pdf

      convert from ppmv to ppmw in gas mixture – dcasto (Chemical)6 Apr 07 09:30
      PPMV in gases is the same a parts per million of moles. Multiply the componets in the gas stream by their molecular wt. This will give grams (or pounds or weight whatever your point of view is). Sum up all the weights and get the mass. Take each componet mass divided by total mass to get coponet fraction including ppm wt.
      OR take the total stream average mole wt and get the total mass of the stream (you can also use spgr). Then take the mole wt of the componet that you want to find ppmwt and multiply by its mole wt. Divide the compnet mass by the total mass of the stream. This is an estimate and works very well for ppm conversion.
      http://www.eng-tips.com/viewthread.cfm?qid=183495

      CONVERTING ATMOSPHERIC POLLUTANT CONCENTRATIONS
      http://www.eng-tips.com/faqs.cfm?fid=1205

      • Further research finds the following:

        A new estimate of the total mass of the atmosphere is 5.1361×1018 kg for the annual mean, corresponding to a mean surface pressure of 984.43 mbar. It ranges from 5.1352×1018 kg in January to 5.1371×1018 kg in July, owing to the annual cycle in global water vapor which has an amplitude of 1.0×1015 kg (0.2 mbar). The total mass of dry air is estimated to be 5.123×1018 kg (or 981.9 mbar).

        Now, the molecular mass of dry air is 28.97 (source), but the atmosphere also contains water. From the data given above, this reduces the average molecular mass to 28.94.
        SO … with the formulas I gave above, using these revised figures of atmospheric mass (5.1361e15 tonnes) and atmospheric molecular mass (28.94) gives us a conversion factor of 2.13 Gtonnes C/ppmv of atmospheric CO2, which I’ll use for my future calculations.
        Thanks for pushing me to investigate this fully,
        Best regards,
        w.

    • Samuel C Cogar January 28, 2015 at 9:38 am Edit

      Willis Eschenbach January 27, 2015 at 5:28 pm

      The difference is that you’ve confused ppmv, parts per million by volume, with ppmw, parts per million by weight.

      Willis, if I am confused ….. then what about Ferdinand, to wit, your original statement followed by his response:

      Cumulative emissions are the sum of all human emissions translated from GtC fossil fuel use to ppmv (factor 2.12 GtC/ppmv)

      And then I found this conversion rate which is different than both the above, to wit:

      The standard conversion rate is 2.13 Gt C is equal to one ppmv CO2

      The variation in these depend on the total mass of the atmosphere, which is not exactly known. See the marvelous Physics Factbook for the values.
      I ran the numbers using 5.15e+15 tonnes as the atmospheric mass. If you use 5e+15, or 5.3e+15, all numbers that I’ve found in the literature, you’ll get the range of answers above. Any one of them could be correct.

      One thing, Willis, I can not figure out where you got your converted rate of “1.55E-06 ppmw”.

      Thats one ppmv times the difference in weight between the average molecular mass of the atmosphere and the molecular mass of CO2. From memory I took the molecular mass of the atmosphere as 28.3, but looking it up I see I should have used 28.97. In any case, since CO2 is much heavier than air, the one part per million by volume is heavier than one millionth of the weight of the atmosphere. So the correct number should have been
      1e-6 * 44 (molecular mass CO2) / 28.97 (molecular mass air) = 1.52e-6
      Best regards,
      w.

    • Willis Eschenbach January 28, 2015 at 10:57 am

      Further research finds the following:

      A new estimate of the total mass of the atmosphere is 5.1361×1018 kg for the annual mean, corresponding to a mean surface pressure of 984.43 mbar. It ranges from 5.1352×1018 kg in January to 5.1371×1018 kg in July, owing to the annual cycle in global water vapor which has an amplitude of 1.0×1015 kg (0.2 mbar).

      Now that is interesting, …. that the atmospheric H2O vapor quantity also has a bi-yearly cycle (January to July) …. just like the atmospheric CO2 quantity has its bi-yearly cycle (September to May).
      Huumm, it appears to me that maybe the air temperature is driving the bi-yearly “H2O cycle” and the water temperature is driving the bi-yearly “CO2 cycle”

      • Samuel,
        I suppose that is indeed a matter of temperature: the NH has a higher seasonal temperature swing than the SH, due to more land (and water evaporation by plants?). Which makes that the global temperature is ~1°C higher in the NH summer than in the NH winter. CO2 goes opposite: lowest levels in summer, highest in winter, again because of more land and thus more vegetation absorbing CO2 in the NH summer, as can be seen in the opposite CO2 and δ13C changes:
        http://www.ferdinand-engelbeen.be/klimaat/klim_img/seasonal_CO2_d13C_MLO_BRW.jpg
        It would be interesting to see the differences in water vapor over the seasons between the NH and the SH…

      • Ferdinand, your above graph appears to be little more than the product of your “creative” thinking ……. and me thinks you should clean-it-up before anymore touting of its scientific value.
        First of all, you titled your graph as being “average seasonal CO2” yet there is nothing “seasonal” about the plotted CO2 data given the fact you “zeroed”each years’s data to begin on January 1 of each “calender year” which has nothing whatsoever to do with seasons or seasonal. The seasons begin and end with equinoxes and solstices …. so “zero” yor data with one of them, …… or least denote all four (4) of them on your graph.
        Of course, that brings up the 2nd and critical pronlem with your graph which renders it pretty much FUBAR. And that is, … no where does your plotted CO2 data agree with factual reality or the Mauna Loa record. ftp://aftp.cmdl.noaa.gov/products/trends/co2/co2_mm_mlo.txt
        Atmospheric CO2 doesn’t start its bi-yearly decrease until post-mid May …. and it doesn’t start its bi-yearly increase until post-September 23rd.

        CO2 goes opposite: lowest levels in summer, highest in winter, again because of more land and thus more vegetation absorbing CO2 in the NH summer,

        Your above mimicry does not comport with the biology of the seasonal changes in the Northern Hemisphere.
        The “greening” of the vegetation in the NH begins in mid to late January in the southern latitudes when the average daily surface temperatures begin increasing ….. and progresses slowly northward to mid June in the far northern latitudes.
        Thus, as the increase in the “greening” of the vegetation progresses slowly northward there is also an accompanying increase in the absorption of atmospheric CO2 which is required for said “greening” photosynthesis. But, even though said increase in “greening” and said increase in “CO2 absorption” began in late January and progressed slowly northward, …… the decrease in atmospheric CO2 did not actually begin until mid-May, …. which is like 3 ½ months after the “greening” started.
        And, even though there is a 3 ½ months delay between the aforesaid “start” cycles …. the novices and the miseducated have been nurtured to believe that the aforesid increased “greening” is the direct cause of the aforesid decreased CO2. But, there is a prooblem with that nurtured belief.
        Said “greening” of the vegetation in the far northern latitudes begins decreasing and/or terminates by late July …..and quickly progresses southward toward the southern latitudes and has pretty much terminated by late August or first of September ….. whereas the atmospheric CO2 continues its decline until after September 23rd … and oftentimes to early October, …. which is like 1 month after the “greening” stopped.
        But the really big problem with the aforesaid “nurtured belief” is the fact that the Springtime increase in average daily surface temperatures, ….. as they progress slowly northward to the far northern latitudes to cause said increase in “greening”, ….. they also “trigger” an horrendous increase in the biological rotting and/or decomposition of dead biomass which causes the emission of copious amounts of CO2 into the atmosphere …. which are probably close to being equal to or greater than the “rate” at what the “greening” is absorbing from the atmosphere.
        So there, Ferdinand, is a quasi example of your touted “mass balance” in real-time action.

      • Samuel,
        The bi-yearly cycle is mainly a combination of four opposite fluxes: warming and cooling ocean surface, vegetation decay and vegetation growth.
        I did zero everything in January to show the net effect that repeats every year, without the overall trend, but including the yearly trend, that is why I included January of next year. That shows what is measured nothing less, nothing more. But that does skew the seasonal changes somewhat. If you want, you can detrend that by subtracting a linear 2 ppmv/year increase.
        In short: the CO2 movements are dominated by vegetation in the NH. That these aren’t 100% synchronized with the start of the seasons everywhere (which shifts with latitude in spring) is a matter of relative fluxes: vegetation decay has a maximum in fall, but goes on all winter and increases in spring and summer, but is overwhelmed by the spring and summer growth. Ocean fluxes are opposite to vegetation fluxes…
        All what I have done is showing the net result of all the CO2 fluxes in the NH atmosphere at different heights (Mauna Loa lags Barrow somewhat and is more smoothed), which show that the oceans are not the dominant force of the seasonal or 2-3 year variations, as you seem to prefer…

      • (Ferdinand) I did zero everything in January to show the net effect that repeats every year,

        Ferdinand, zero everything at September 23rd and see what you get.

        (Ferdinand) you can detrend that by subtracting a linear 2 ppmv/year increase.

        If your plotting the bi-tearly cycles why the ell did you include .. a linear 2 ppmv/year increase? Did you need it to prove your assertion?

        (Ferdinand) In short: the CO2 movements are dominated by vegetation in the NH. That these aren’t 100% synchronized with the start of the seasons everywhere (which shifts with latitude in spring) is a matter of relative fluxes:

        Ferdinand, that “weazelworded” dog won’t hunt. Do you “relatively flux” them to where you need them ……. or what. Besides, I’ve never witnessed any “relatives flux”.
        Ferdi, the start of the seasons everywhere (which shifts with latitude in spring) ….. IS the start of the vegetation growth everywhere.

        (Ferdinand) vegetation decay has a maximum in fall,

        YUP, shur nuff, Ferdinand, …. and all frogs have wings …. for flying their tadpoles to differennt puddles of water during drouth periods.
        Ferdinand, the fall season of each and every calender year begins “hot n’ dry” in late August and continues into November being “cold n’ wet” ….. but the “cool down” always precedes the “wet down” ….. and only rarely is it ever “warm n’ wet” during the fall season.
        To wit, educate yourself Ferdinand. even if you close your mind to it afterwards:

        Proper Storage Temperatures for USDA Commodities
        Dry Storage – Store dry foods at 50°F for maximum shelf life. However, 70°F is adequate for dry storage of most products. (HOT N’ DRY WON’T DECAY)
        Refrigerated Storage – Refrigeration increases shelf life of most products. Most importantly, refrigeration slows bacterial growth. Maintain refrigerated storage spaces at 32-40°F. (COLD N’ DRY nor COLD N’ WET WON’T DECAY)
        Freezer Storage – Freezers should be used to store frozen food when it is received. Maintain freezer storage spaces at 0°F or below. (FROZEN SOLID WON’T DECAY)
        http://www.cde.ca.gov/ls/nu/fd/mb00404.asp

        (Ferdinand) but is overwhelmed by the spring and summer growth.

        PHOOEY, the majority of spring growth is accomplished via sugars stored in the seeds or roots. The vegetation can’t start sucking in CO2 until they have fully formed leaves with functioning stomata.

        (Ferdinand) Ocean fluxes are opposite to vegetation fluxes…

        Ferdinand, that was a “HALF-TRUTH” and should have been stated thustly: “Southern Hemisphere ocean fluxes are opposite to Northern Hemisphere vegetation fluxes” ….. and the SH ocean fluxes overwhelms the NH vegetation fluxes.

        (Ferdinand) (Mauna Loa lags Barrow somewhat and is more smoothed), which show that the oceans are not the dominant force of the seasonal or 2-3 year variations, as you seem to prefer…

        Double PHOOEY, …. Ferdinand, …. it is immaterial whether or not Mauna Loa Hawaii (@ 19.4 N latitude) …. lags Barrow, Alaska (@ 71.3 N latitude) when the primary driver of both the average 6 ppm bi-yearly cycling and the average 2 ppm yearly increase in CO2 is the Southern Hemisphere oceans (@ equator to 75 S latitude).
        Ferdinand, the scientifically measured ….. yearly maximum atmospheric CO2 ppm occurs during a very narrow “window” during the middle of May of each calender year …… which you have NEVER offered a sensible or logical explanation for …. in all your toutings of fluxes, biases, estimates, guesstimates, “fuzzy math” average surface temperatures, etc., etc.
        And the same goes for the yearly minimum atmospheric CO2 ppm that occurs during a very narrow “window” at the end of September of each calender year.
        Ferdinand, you have to explain the above stated NON-anomaly of a “narrow window” ….. before you take a “leap-of-faith” jump into a bucket of “junk science” tripe n’ piffle and then attempt to worm your way out of it.
        Ignoring them will not cause “facts of the natural world” to disappear.

  53. @Willis “How could they lose CO2 but not air?”
    Gosh, I wonder? However could that happen? Possibly because CO2 is significantly more soluble than O2 meaning more CO2 trapped inside the liquid ice bubble will be absorbed by it and gradually leach out into the ice at a vastly greater rate than O2. That’s one obvious explanation that comes to mind. Perhaps if you weren’t so busy bashing Jaworowski and actually thought about things a little more, you might have considered that.

    • Richard, we have looked very deep into all aspects of what Jaworowski said, but couldn’t find anything in his defense. Like that there is no difference between gas age and ice age. Even if you don’t know much about ice cores, it is quite easy to understand that the pores remain much longer open than the ice layers which staple above each other and thus that the average gas age is a lot younger than of the ice.
      Even if CO2 was leaching into the liquid-like layer (some 5 molecules thick at -20°C) on the ice, CO2 measurements are performed by grating the ice samples at low temperatures and high vacuum over a cold trap to remove any water from the ice (including CO2) and water out of the air. Thus if air was leaking first and CO2 was hiding, at measurement time that would lead to increased CO2 levels, not to a 30-50% loss as Jaworowski assumes (without a shred of evidence).
      The latest technique is by sublimating all ice just under the melting point and freezing CO2 out cryogenically, then measuring over a mass spectrometer so that besides the absolute levels the isotopic composition can be measured too. That also effectively destroys clathrates which may have remained in the ice. But even so, the grating technique shows the same CO2 levels as the sublimation technique, thus not much CO2 is hiding anywhere. See for a description:
      http://www.awi.de/de/forschung/fachbereiche/geowissenschaften/glaziologie/techniques/high_precision_d13c_and_co2_analysis/
      That was used for the full length of the Dome C ice core (800,000 years back in time).
      Thus sorry, what Jaworowski said about ice cores is physically impossible and not backed by any evidence in the field…

    • Well, Richard, I came here to answer your stupidly condescending comment by saying that Ferdinand had already discussed it, and found that he’s much kinder than I am, so he has repeated and expanded on his previous explanation.
      Let me suggest that you lay off of the sarcastic asides like “Gosh, I wonder? However could that happen?”. All you’ve done by such comments is shown that not only are you not following the story, and not only do you truly not understand the processes used and the physics in play, but that you want to be an arrogant jerk about it …
      w.

  54. @Ferdinand
    I can’t say I’ve ever heard Jaworowski argue that the ice-core underestimates past CO2 by 30-50% but he does cite measurements of the surface-snow from other researchers showing that the snow underestimates atmospheric CO2 by 20-50%.

    • Richard, if CO2 was really 30-50% lower in snow than in the surrounding air, I would expect the same problem further down where the snow is compacting to firn and ice. Snow crystals are extremely open: it is a little ice and a lot of air. If anything one would expect higher levels of CO2, not lower, depending of temperature and thus amount of liquid-like water at the surface of the ice crystals…

  55. Regarding Jaworowski’s take on the different extraction methods, regardless whether you think it is possible or not, there is an interesting graph he cites from one researcher comparing the melt-extraction measurements with the older dry-extraction measurements and the former shows CO2 over 900ppmv. CO2 being absorbed by the liquid bubble seems reasonable to me on the face of it and it would explain the disparity in the CO2 levels between these two extraction methods. Jaworowski also touches on how higher measurements were removed (somewhat inappropriately) by Neftel 1982. There is so much intrinsic uncertainty in paleoclimate ice-core data and so many potential fractionation issues I think it would be unwise to assume they are an accurate representation of ancient CO2 levels, especially when direct measurements of the surface-snow show that that the snow underestimates atmospheric CO2 by 20-50%.

    • chipstero7, there are known issues with Greenland ice cores: much more sea salt deposits, including carbonates than in Antarctic cores, but more important frequent highly acidic ash deposits from nearby Icelandic volcanoes. That is what gives in situ reactions which increase the CO2 levels. With the wet method, all ice is melted and the CO2 extracted with vacuum, but the acid-carbonate reaction goes on, thus increasing the CO2 release over time. That is the reason that the wet method is completely abandoned and CO2 levels from the Greenland ice cores are not reliable and not used.
      Antarctic cores don’t have that problem – or at least to a much lower extent.
      Neftel was one of the first to drill ice cores (at Siple Dome) and as with any new technique, they had mechanical troubles, including cracks in the ice, where drilling fluid was found in the cracks. At these parts of the ice core, huge differences were found in multiple samples. Neftel rejected the high levels, as these were highly probable from the contamination. I should have rejected all measurements of these parts of the ice as unreliable… Later there was a new core drilled at Siple Dome (Friedly?) which didn’t show any extremes…
      Even in the most recent Dome C core, there is one part of the core with extreme values, as they had to inject alcohol to loose the drill which was stuck in ice near maximum depth.
      I don’t know of any underestimate of CO2 in snow near the surface (citation?), Etheridge e.a. measured snow and firn top down from the surface and didn’t find a difference near the surface:
      http://www.ferdinand-engelbeen.be/klimaat/klim_img/law_dome_firn.jpg
      I know, Jaworowski did write a lot of pertinent questions regarding ice cores of the early days, but Etheridge e.a. repeated all his objections, researched each of them and provided a definitive answer as found in their fieldwork. That was in 1996. Since 1996 ice core research didn’t stop, but for some reason Jaworowski did stop since 1992 to accept any new research which refuted his objections…
      See further:
      http://www.ferdinand-engelbeen.be/klimaat/jaworowski.html
      Thus in my informed opinion, ice cores give reliable information about CO2 and other gases from the past, be it in smoothed form, where the resolution depends of the accumulation rate for each core.

  56. Issues that concern me include, in a few words:
    (1) diffusion of CO2 through the ice after ‘sequestration’ (tends to smooth out variations),
    (2) differential diffusion rates due to the variable densities and porosities of firn/ice,
    (3) non-linearity and discontinuity in the snow/ice deposition rates,
    (4) variability of temperatures (even at <0° C the absorbtion of CO2 by water/ice is temperature dependent),
    (5) linear resolution problems in sampling ice for CO2,
    (6) identification/dating of the 'layers' of snow/firn/ice.

    • And, no means of verifying whether the measures they take to “compensate” for all this really work. Proponents of the ice core measurements are long on assertion, but short on actual proof.

    • tadchem,
      (1) immeasurable small in 800,000 years old ice over 8 glacial/interglacial transitions.
      (2) calculated and measured in firn, immeasurable in ice.
      (3) agreed, but only important for the gas age – ice age difference thus the timing and the resolution, thus peak detection, but not important for the CO2 average over the resolution period.
      (4) highly dependent of temperature: very small at -20°C, non-existent below -32°C. Vostok, Dome C and other central high altitude ice cores are average -40°C. Not important at measuring time as almost all water-like layers are removed under vacuum, including any CO2 which would hide there.
      (5) resolution depends of accumulation rate and is from 10 years (Law Dome summit) to 600 years (Vostok), but as many ice cores with extreme differences in accumulation rate, temperature and resolution overlap each other, not a real problem. The current increase of 110 ppmv over 160 years would be noticed in every ice core if it happened anywhere in their time span.
      (6) agreed, but only important for the gas age – ice age difference thus the timing and the resolution, thus peak detection, but not important for the CO2 average over the resolution period.

  57. Sigh.
    I’m going to have to find another hobby. That exchange between David Socrates and Bart was mind-numbing, and because they did it all with REPLIES in a THREAD, one couldn’t even jump in to indulge in some harmless flaming and name calling and trolling myself without scrolling back a mile or so to find the first reply button. Which (no doubt) saved me from myself, big time.
    So, since there is no time, let me elevate myself (probably undeservedly) to the role of a referee and sum up:
    Between Ferdinand and Bart — you two should really not argue as much and spend more time listening to each other. Bart, Ferdinand’s mathematics looks like it is very carefully considered, is soundly based on measurements, and AFAICT he is competent as a mathematician. Ferdinand, Bart’s mathematics looks like it has been carefully considered, is based on actual measurements. I have no particular reason to think that Bart is incompetent in ODEs of first order based on what he has said or published over the years.
    You two differ in the models you use to explain the same observations, which is just fine! There could easily be more than one model that adequately explain the same observations in a highly multivariate systems projected in to a much lower dimensionality, and all of the models that work (so far) could still be wrong! So showing a tiny bit of respect for the other’s point of view would work better than name calling (especially calling each other liars by statistics when you both are obviously — to me — quite honest and just disagree). Keeping the discussion purely quantitative would be even better, although I admit that it is not going to resolve the problem when both of you have models that sort-of-work within the unknown parameters of stuff like “what shifts the equilibrium” and the fact that the differential systems are possibly absurd oversimplifications of what is actually going on.
    So shake hands, blushing slightly, and go buy each other a few beers or something to take the heat out of the discussion.
    Bart and Socrates — Bart the winner on points but that is almost entirely because Socrates has taken the handle Socrates. Every single person I’ve met on the Internet who calls him or herself Socrates shouldn’t. It is sort of like the ultimate anti-label. Of course Socrates could be David’s last name (just like Bart could have named himself after Bart Simpson) but if it isn’t, oh my.
    Oh, and regarding content Bart wins with a knockout, because Socrates “argument” was stunningly math free. Bart may or may not be correct in the sense that his model (however well it works on selected past data) proves to have predictive power in the future, which is the only sense in which any sort of model matter!, but at least his arguments are clear and easily understood by anybody who understands first order ODEs or small systems of 1st order ODEs. I do. (So does Ferdinand, Bart, which is why you shouldn’t be quick to call his comments either misguided, malicious, or ignorant.) Socrates I suspect wouldn’t know a 1st order ODE from an ordinary algebraic equation as he frequently confuses the concepts in his discussion.
    Socrates, please — and I mean this in the very best of ways — learn some math before you engage in a QUANTITATIVE discussion on math. I personally don’t care if Salby’s model is correct or incorrect, but you can’t argue for it by saying other people (whoever they are) believe it or disbelieve it, or by referring to it obliquely, as a four or five sided discussion with the principles absent and the things being discussed all elsewhere is a cosmic waste of time. Start a thread on “Salby’s error” or something and tune us all in, with meaningful quantitative support.
    I know a dragonslayer who likes to usurp greek philosopher names in vain as well. You are sounding way too much like him, only a mirror image with the opposite point of view. How about dropping the handle and actually being reasonable?
    rgb

    • Thanks for your input, RGB. In the end, being right is not what carries the day. It is in whether or not you appeal to the audience.
      Ferdinand is a very nice fellow. I hate taking him on. I hate that he is so very wrong. But, he is.
      His content is almost all assertion, not math. He does not have a strong, mathematically based background in feedback systems, which is essential to understanding what is going on.
      Can we agree that the CO2 data versus temperature anomaly is, within the constraints of noisy and error prone measurements, essentially a perfect match to
      dCO2/dt = k*(T – Teq)
      ?
      This equation tells you everything you need to know to estimate a value of CO2 in the atmosphere at any time since 1958, given only the starting value, and the temperature record since.
      You cannot add a significant portion of human inputs to the result – if you do, you have to rescale k to fit it in. But, when you rescale k, the variations no longer match. You can futz around with it, and maybe come up with a scenario in which human inputs provide perhaps up to 30% of the total, but it’s a strain, and there is no reason for it.
      Occam’s razor says that the simplest explanation is likely the right one. The explanation that says the equilibrium level of CO2 in the atmosphere is regulated in line with temperature dependent boundary conditions, that human inputs are attenuated severely by that regulatory system, and that changing boundary conditions have driven the observed rise, is the simplest and most straightforward explanation. Any other explanation is a balancing act requiring an unlikely confluence of finely tuned responses.
      Ferdinand insists that you can add in significant human forcing if you use a model in which CO2 variation is directly proportional to T. In this way, he eliminates the slope in T from the dCO2/dt equation, and thereby allows human forcing to carry the day.
      This does not work. The variations in dCO2/dt match the variations in T, which says there is a 90 deg phase lag between T and CO2. Ferdinand dismisses the phase response as nothing important. He is completely and utterly wrong on that.
      http://i1136.photobucket.com/albums/n488/Bartemis/dTleadsdCO2_.jpg
      Figure: It is easy to see the 90 deg (1/4 cycle) phase lag in the plot of dT/dt and dCO2/dt. The data here are from GISS.
      It is a well known result in systems theory that a 90 deg phase lag means a -20 dB/decade gain. This is a result of the storied Bode Gain-Phase theorem, which I would go as far as to say is the fundamental theorem in systems control theory. You get a 90 deg phase lag for every -20 dB/decade rolloff in the response. Formal proof requires assumption of a minimum phase linear system (no poles or zeros in the right half plane) but non-minimum phase systems are easily recognizable in the large phase lags they produce, and the system with which we are concerned is observably not one of them.
      Design engineers know that they must have an open loop gain rolloff much less steep than -40 dB/decade in order to have a stable closed loop system with good phase margin. Otherwise, they get near -180 deg phase lag in the open loop which produces instability when the loop is closed. In very simple and direct terms, 90 deg of phase lag across the spectrum is equivalent to -20 dB/decade gain, and -20 dB/decade gain is the response of an integrator. A natural system with continuous response which displays a 90 deg phase lag over all frequencies is an integrator. There is no way around this.
      For our data set, it is possible that the integration stops after some time interval (below some frequency, the -20 dB/decade response could level off), due to a long term feedback asserting itself. But, it does not matter. For all practical purposes, the data we have since 1958 displays 90 deg phase lag across all observable frequencies. It is not possible to have a time constant of lesser duration than at least 3 times the record length, and that’s being generous – more likely, you’d need 10X the record length to get data so completely devoid of any noticeable phase distortion.
      As a result, the trend in T must integrate into the trend in dCO2/dt. And, the existence of that trend obviates significant contribution from human inputs. Human inputs also have a trend. The trend is already accounted for, and there is little to no room for more.
      Human emissions are accelerating. The rate of change of CO2 concentration has stalled, in lockstep with the temperature stall.
      http://i1136.photobucket.com/albums/n488/Bartemis/CO2_zps330ee8fa.jpg
      The plot above was made a couple of years ago. Since then, CDIAC has not updated the emissions numbers, and I can no longer compare them, but I have no doubt the discrepancy is even greater now.
      These two items, the stall in concentration even as emissions accelerate, and the 90 deg phase lag indicating the necessity of the dependence of CO2 on the integral of temperature, tell us without a doubt that the model which says CO2 levels are due to human forcing is broken beyond repair.

    • rgb,
      Sometimes I am wondering too why I didn’t work more on one of my (too) many other hobby’s instead of reacting here (and a lot of times the other way out against “warmistas” in our local newspapers)…
      I am not light in accusing someone of lying, but if one is showing a graph, again and again, that suggests some problem which isn’t a problem at all, and I have warned him five times, then even my patience is gone.
      Have a look at a similar graph as Bart’s with the same data, with the proper scales and units for the emissions, increase in the atmosphere and sink rate, all yearly averages:
      http://www.ferdinand-engelbeen.be/klimaat/klim_img/dco2_em2.jpg
      The yearly increase of CO2 in the atmosphere in the period 1994-2011 was rather flat, as was the period 1975-1994, be it at a lower level. According to Bart that is a sign that the increase in the atmosphere is not related to the emissions. Even not if the emissions still are twice the increase in the atmosphere and the variability is in the sink rate, not the source rate…
      The sink rate doesn’t even depend of the momentary emissions, the sink rate depends of the pCO2 difference between atmosphere and mainly oceans, the latter is slightly influenced by temperature (~8 ppmv/K, according to the literature: 4-17 ppmv/K).
      In my professional life (chemical engineering, later process automation) I did mostly practical work: upscaling of processes from laboratory to factory and later mostly implementing new computer systems for the chemical processes (batch, later continuous). That makes that all my theoretical process knowledge is very rusty and far less than what Bart knows in theory, but that my practical knowledge of processes is far superior than his.
      Thus our repeated (over the past years) discussion is mostly of theory against practice and theory against observations…
      The latest one is somewhere under the long list of answers and answers on answers between Bart and Socrates here.
      The problem is that Bart has one point for his theory: a nice fit of the variability of CO2 and temperature. From that he theorizes that the slope of the CO2 rate of change also is caused by temperature (by arbitrary matching two straight lines…), but the problem is that the variability is caused by the fast reaction of (tropical) vegetation to temperature changes, while the trend is not from vegetation: that is a proven, increasing sink for CO2…
      Further, Bart doesn’t accept any observation which shows that his theory can’t be true. That makes real discussions quite impossible.
      Take his main formula:
      dCO2/dt = k*(T – Teq)
      there is a theoretical match with the real CO2 increase since Mauna Loa started, but from a process viewpoint impossible: a small, sustained increase in temperature will give a continuous increase of CO2 in the atmosphere until eternity, without any feedback from the increased pressure in the atmosphere. Further, try to match that for other periods of the earth, even in the recent past: 1945-1975 shows a (small) cooling, but CO2 goes up, so k then is negative or one need to shift Teq down a lot to still have a positive k… Not even to mention the k needed to give the dCO2/dt for 5000 years of glacial – interglacial transitions. But Bart – of course – reject ice cores as reliable…
      Then his demonstration of the CO2 lag:
      Any increase of temperature of the oceans (vegetation in general gives the opposite reaction) increases the CO2 outflux of the oceans. But that happens with a lag, here for 1 K temperature increase and 18 ppmv/K shift in equilibrium (somewhat overdone):
      http://www.ferdinand-engelbeen.be/klimaat/klim_img/upwelling_temp.jpg
      That simply means that after some time, the in/out fluxes of the oceans are back in equilibrium at a higher level of CO2 in the atmosphere. No continuous inflow for a sustained increase in temperature.
      If you have a sinusoid of temperature changes that gives a sinusoid in CO2 changes with a lag of 90 deg. That is what Paul_K figured out:
      http://bishophill.squarespace.com/blog/2013/10/21/diary-date-murry-salby.html
      Page 2 of the comments, fourth comment…
      Thus while T changes causes CO2 changes, the frequencies of CO2 lag T with 90 deg.
      If you take the derivatives of both they shift back 90 deg. in time, but still the variability of dT/dt causes the variability of dCO2/dt with a 90 deg. lag.
      Now Bart compares the variability of T with the variability of dCO2/dt. Of course that shows a perfect match, as dCO2/dt has made a 90 deg, shift back in time thus now is synchronized with the variability in T, but that has no physical meaning at all…
      What I have read of the Bode Gain-Phase theorem is that it is for a one-variable closed loop system with a huge feedback. The T-CO2 relationship is mainly an open loop system without much feedback of CO2 on temperature.
      I have done some experiments with Matlab and don’t see any influence in phase distortion (always 90 deg.) for any frequency of the integration of a linear increasing addition to a sinusoid of different frequencies, only the amplitude of the variability gets lower with higher frequencies.
      Thus given the violation of all known observations by Bart’s theory, in my informed opinion, his theory is wrong…

      • It’s the same graph, Ferdinand. Your emission rate has a slope – emissions are accelerating. Your atmospheric data has no identifiable slope since the “pause”. This is creating a large divergence in accumulated emissions versus atmospheric concentration. No game playing with the scales of the graph changes that.

      • “Now Bart compares the variability of T with the variability of dCO2/dt. Of course that shows a perfect match, as dCO2/dt has made a 90 deg, shift back in time thus now is synchronized with the variability in T, but that has no physical meaning at all…”
        You all see, Ferdinand does not think phase has any meaning.
        It does. If he cannot match the phase, then his model is wrong.

      • “What I have read of the Bode Gain-Phase theorem is that it is for a one-variable closed loop system with a huge feedback.”
        The Bode Gain-Phase Theorem applies to any rational, minimum phase transfer function. There can be no argument against this.

      • “I have done some experiments with Matlab and don’t see any influence in phase distortion (always 90 deg.) for any frequency of the integration of a linear increasing addition to a sinusoid of different frequencies…”
        We’ve been over this on another thread. Your “simulations” are not valid.

      • BTW: I am not the one who made things ugly with accusations of lying, but I will reply in kind. Stop lying with the fallacious “mass balance” argument, which you know does not prove anything.

      • Bart,
        You can’t violate a mass balance. No matter how much more natural CO2 you try to inject into the atmosphere, even if it was a 4-fold, the total amount of injected natural CO2 goes down into natural sinks + a part of the human emissions. The smaller the ratio human/natural, the smaller the human contribution in the sinks (or you violate the equality principle of CO2, whatever its origin). The total increase still is caused by the human emissions…

      • Ferdinand,
        ..
        Since Bart doesn’t accpet “mass balance” I do hope that he has someone balance his checking account each month. Otherwise he’d be prone to running up excessive overdraft fees at his bank.

      • It’s not a true “mass balance”. It’s just a fallacious argument. I have punctured it thoroughly above.

      • Bart, you just think you punctured it.
        ..
        You disregard evidence.
        You toss logic to the trash heap
        You ignore Ferdinand’s destruction of your “theory”
        ..
        I’m of the opinion that you can’t dissuade a religious fanatic such as you.
        Enjoy your “religion” the rest of us will practice science.

    • OK, so here is a comment you will neither like. You could ever so easily be completely right, partially right, or completely wrong. Well, probably not completely wrong, but perhaps “mostly” wrong as in missing the key/critical factor.
      Let us just for grins suppose that the atmosphere is dynamically disequilibrated in CO_2. That’s a pretty harmless assumption, since it is manifestly not stationary. Some would even call it a simple observational fact. Second let’s assume, as Bart does, that it is trying to relax to some (possibly mythical) “true equilibrium”. I say possibly mythical because Bart’s equations are in some fundamental sense a mean field theory projecting what is almost certainly complex multidimensional dynamics onto the time axis, where the actual dynamics could be poincare cycles around a fixed point in N dimensions, not relaxation, but because he is linearizing in 1 D it will look like simple relaxation across any sufficiently short time simply because that’s how Taylor series work. Note well that I am not stating that this step is “right” or “wrong” — when modelling a complex phenomenon at some point or another after studying the data one has to choose a model, unless one is using an unbiased but meaningless universal model such as a neural net.
      Bart has selected a model that says “the model is disequilibrated in CO_2 because the CO_2 equilibrium fixed point is determined by the temperature”. When the temperature rises, the equilibrium CO_2 concentration rises, disequilibrating the atmosphere. The sum totality of sources and sinks and feedbacks then conspire to increase CO_2 to relax to the new equilibrium. This relaxation proceeds at a rate k, proportional to the temperature increase relative to the equilibrium temperature.
      This model is not without problems. For one thing, what exactly is T_{eq}, both conceptually and numerically? For another, is T_{eq} itself stationary? It is safe enough to assume (again, really an observable fact) that T(t), the temperature itself is a function of time and is very much not stationary. As a consequence
      dCO_2/dt = k(T - T_{eq})
      basically says that CO_2 will increase without bound as long as the temperature remains above this equilibrium temperature. And what determines this temperature? If it varies independent of T(t) itself, of what use is this equation for projecting the model into the future or the past? How can one compute it, or at least argue that it should have some particular value? If (for example) it is 290 K, cCO_2 should be falling. If it is 280 K, we are cosmically screwed, as cCO2 will increase without bound until the next ice age (according to this formula).
      Since the latter seems (to me) to be more than a bit implausible, it appears (again, to me) that Bart’s model, however well or poorly it fits a (remarkably short) segment of the data from the very recent past, is at the very least incomplete. It contains absolutely nothing to describe some sort of eventual equilibration even if the temperature becomes constant, as long as that constant is higher than T_{eq}. It cannot possibly work in the distant past or the distant future, or else the atmospheric chemistry would be fundamentally unstable, and if there were any sort of greenhouse effect either positive or negative feedback would cause increase or decrease without bound.
      In order to be taken seriously, then, the model requires at the very least T_{eq}(t) and an equation that predicts it (because, as I said, T_{eq} as a universal constant is directly contradicted by billions of years of atmospheric stability against runaway CO_2 concentration). We then are led to a very interesting fork in the road.
      Does this temperature that controls whether or not CO_2 locally increases or decreases in time depend on CO_2 concentration? Does it depend, in any way, on the total amount of carbon in play in the entire system of reservoirs? Does it depend in any way on the details of how those reservoirs and their “capacity” for CO_2?
      Again, one has to be very careful with one’s answers. Basically, the answer to one or more of these had better be yes or one cannot prevent the implicit runaway catastrophe in the model. That is because there must be, somewhere in the system of equations, a projection that looks like this:
      dcCO_2/dt = \kappa (cCO_{2,eq}(t,T_1, T_2,x,y,z...kitchen sink) - cCO_2)
      In words: There must exist a non-stationary equilibrium CO_2 concentration for the atmosphere. This equilibrium concentration can and probably does depend on many things — time, air temperature, ocean temperature, ocean state (as Bart himself has noted) and yes, sure, one of those parameters could well be “total carbon in play in the system” and hence be changing due to anthropogenic input.
      If the current CO_2 concentration is less than the equilibrium, its rate of change is positive, driving it towards equilibrium. If it higher, the rate of change is negative, driving it back to equilibrium. The system is then locally stable, and unless internal feedbacks in the still omitted equations that determine the time variation of cCO_{2,eq} itself, which very likely is coupled to cCO_2 itself among many other things, are themselves positive in certain ways, it is globally stable as well. I’d obviously argue strongly for global stability because of the finitude of the carbon available in the entire system and the simple fact that while it has varied considerably on geological time, CO_2 has never evolved to either trivial fixed point of all CO_2 in the atmosphere or all CO_2 bound up in reservoirs in a CO_2 free atmosphere (basically, eternal hothouse Earth or eternal snowball earth, in all probability).
      Even this equation, with all of the implicit power of making the equilibrium concentration a function of the kitchen sink as necessary, is probably not sufficient. It really doesn’t allow, for example, for adding a simple, direct bolus of CO_2 to the atmosphere from carbon sources outside of the existing system. It omits us. If we add us back in:
      dcCO_2/dt = \kappa (cCO_{2,eq}(t,T_1, T_2,x,y,z...kitchen sink) - cCO_2) + F(t)
      where F(t) represents the direct addition of CO_2 to the system by means not described in the shift of equilibrium CO_2 concentration and relaxation processes within the system, then the system might be sufficient. Without this, one couldn’t (say) double atmospheric CO_2 overnight and then watch the system relax to equilibrium (the same or a different one).
      To conclude, this latter equation doesn’t mean that Bart’s description is contradicted. For Bart’s description to work locally, it suffices to make cCO2_{2,eq} a function of atmospheric temperature but wait — it already is, obviously, a function of temperature, and beyond that k is just a product of \kappa and the linear term projected into the T dimension from some multidimensional Taylor series of cCO_{2,eq} about a more or less arbitrary point (hence the lack of meaning of T_{eq}).
      I’m not arguing for any particular form for the equilibrium concentration as a function of kitchen sinks. Nor am I taking a position concerning what is the best local projection of this multivariate differential into a restricted subspace of variables, although I do suspect that nearly all such projections will be naive and of limited utility outside of an interval where the linearizations happen to work (which we probably cannot compute a priori, making linearized models nearly useless as predictors as they’ll work — until they don’t). It is just to show that both you, Bart, and you, Ferdinand, can be entirely correct about assertions of agreement of your simple models and just be looking at different projections of the same basic general description of the dynamics.
      This means that there is really no good reason for either of you to call the other a “liar” or even mistaken. What you are arguing about is what terms are the most important, which views of the data are most informative, given highly incomplete, noisy, and in the period over which Bart is looking, boringly monotonic data.
      If I were to participate myself — and I’m not — I’d note that there is a very good correspondence between the rate of increase of atmospheric CO_2 over the last 10-15 years and the anthropogenic CO_2 contribution. A very, very good correspondence. Like, unity. If one assumes we are adding roughly 10 GT/year, that works out to 2 ppm/year, and gee, that’s a very good approximation to the slope of Mauna Loa data. This doesn’t completely contradict a 50 year relaxation rate, but it does mean that we are adding CO_2 at a rate that far exceeds this relaxation rate so that the net concentration is essentially perfectly tracking the additions as we remain far from equilibrium — the F(t) term in the equation above is dominant even if the system is disequilibrated high instead of low because \kappa is too small for us to see more than modulation of the response (what Bart is picking up on).
      My last suggestion is that you two start a thread on this topic alone, independent of the ice core stuff. Put together a complete statement of your position — not just one equation in an obviously incomplete theory — and make sure that there are equations. Too much of the argument above is presenting a graph of this or graph of that where one cannot see where the data came from and what is data and what is a computation and what equation is supposed to be a sufficient explanation. Also, be prepared to defend your equation/hypothesis outside of the range of Mauna Loa. For better or worse, the Mauna Loa data is bo-ring and monotonic, and I can fit a monotonic function on a short interval lots of ways without the fit having the slightest extrapolative virtue. Any model proposed should have at least some virtue across all of HadCRUT4 and all of the combined ice core and ML data. And no omitted detail! If the model is manifestly unstable (e.g. Bart’s) you MUST address this. I’m fine with local linearization, but if it is globally unstable it surely isn’t crazy to require some statement about what nonlinear terms or multivariate behavior will eventually emerge to quench it.
      rgb

      • “Bart has selected a model that says “the model is disequilibrated in CO_2 because the CO_2 equilibrium fixed point is determined by the temperature”. When the temperature rises, the equilibrium CO_2 concentration rises, disequilibrating the atmosphere.”
        Not so much. I am appreciative of others’ objection that temperature alone should not be able to drive the rise which has been observed. What the equation
        dCO2/dt = k*(T – Teq)
        represents is a flow rate from some source which is merely modulated by the departure of temperatures from their nominal range relative to temperature anomaly Teq.
        “In order to be taken seriously, then, the model requires at the very least T_{eq}(t) and an equation that predicts it (because, as I said, T_{eq} as a universal constant is directly contradicted by billions of years of atmospheric stability against runaway CO_2 concentration).”
        There is no requirement that Teq be a universal constant. It is merely the offset parameter of a linearized function which holds within some neighborhood (of space and time) of a particular state of the system. It is remarkable, to me (but then, we humans tend to have a perception of time tied to our brief lifespans) that a constant parameter value holds up quite well over the past half century. But, there it is.
        “…and yes, sure, one of those parameters could well be “total carbon in play in the system” and hence be changing due to anthropogenic input.”
        Atmospheric CO2 is evidently a temperature modulated process. Temperatures have been rising roughly linearly. Rates of human emissions have also been rising roughly linearly. If rates of human emission were being modulated by temperatures, that would produce a quadratic rise in the rate of change of atmospheric concentration.
        However, the rate of change of atmospheric concentration has been rising only linearly, which indicates that the rate of change of whatever process is driving the rise of atmospheric CO2 is essentially constant.
        The rate of human emissions is not constant. Hence, this is not the process which is driving atmospheric concentration
        “…If we add us back in:”
        If we add emissions in at this level, there is little to no room for it. The temperature modulated process already accounts for the trend in the rate of change of atmospheric CO2. Human emissions would add to that trend, making it too large.
        The effect of human emissions is analogous to this system:
        dCO2/dt = (CO2eq – CO2)/tau + H
        dCO2eq/dt = k*(T – Teq)
        In this system, CO2eq is the “instantaneous equilibrium”, or current attractor if you wish to eschew that somewhat oxymoronic phrase, for CO2 in the atmosphere set by the boundary conditions, most likely dominated by the atmosphere/ocean interface. CO2 is then the response, including all sink activity, to the imposed boundary condition.
        In this system, CO2 tracks CO2eq, with increasing fidelity as tau becomes short. The contribution from H to CO2 is approximately H*tau which, again if tau is short (aggressive sink activity) is small.
        This is a description of a very ordinary feedback regulation system. There is nothing at all exotic or unlikely about it.
        “I’d note that there is a very good correspondence between the rate of increase of atmospheric CO_2 over the last 10-15 years and the anthropogenic CO_2 contribution. A very, very good correspondence.”
        A very, very poor correspondence, especially in the last 10-15 years. Human emissions are accelerating. Atmospheric concentration is not.
        This isn’t even debatable. Look at the plot.
        http://i1136.photobucket.com/albums/n488/Bartemis/CO2_zps330ee8fa.jpg
        Look at Ferdinand’s plot, if you like. He has artfully fitted the two series using a least squares, affine map, but he still cannot cover up the fact that human emissions are accelerating, and atmospheric concentration is not.
        “I’m fine with local linearization, but if it is globally unstable it surely isn’t crazy to require some statement about what nonlinear terms or multivariate behavior will eventually emerge to quench it.”
        We do not have to know that to make conclusions about the current era, and determine that the temperature dependent forcing of atmospheric CO2 is dominant in the modern era since 1958.

      • “…If we add us back in:”
        Moreover, adding in human emissions at this level produces the wrong phase response. The bumps and burbles in T lead those in CO2.

      • RGB,
        Herewith I retract my statement that Bart is lying with his graph. To lie, one need to do that deliberate, while in this case Bart doesn’t even understand that his graph is misleading…
        One small remark: the translation of 1 ppmv CO2 is about 2.13 (to 2.15) GtC, depending of what you take as water vapor level in the atmosphere. That makes that the ~9 GtC/year emissions translate to ~4.2 ppmv/year of which ~2.15 ppmv/year remains in the atmosphere.
        My stake in this case is at one side the very good overall fit between temperature and CO2 levels in the past of ~8 ppmv/K which looks like a simple linear first order process, even if a lot of underlying processes might be far from linear. Here for Vostok:
        http://www.ferdinand-engelbeen.be/klimaat/klim_img/Vostok_trends.gif
        Where the temperature anomaly is for the water vapor catch area (most of the SH oceans) which was disposed as snow and compacted to ice, based on d18O changes.
        The largest differences with the trend are mainly from the different lags of CO2 after temperature changes: 800 +/- 600 years during a glacial-interglacial transition and several thousands of years for the opposite transition for which is not corrected.
        On the other side the very good fit of the total of all human emissions ever released and the increase in the atmosphere, or in fact the sink rate, as any increase in the atmosphere is met with an increase in sinks beyond the usual (pre-industrial) dynamic equilibrium of (seasonal and short term) natural CO2 inputs and outputs.
        Taking the two above relationships into account, the CO2 equilibrium can be calculated as a simple function of temperature:
        CO2eq = CO2orig + 8 * (T – To)
        Where CO2orig was ~295 ppmv and To was ~0.8°C lower in 1850.
        Then we can calculate the increase in the atmosphere by subtracting the calculated sink rate for an e-fold decay rate of ~51 years (based on the measured average sink rate for the current CO2 pressure above CO2eq for the current temperature) from the emissions. That gives following curve:
        http://www.ferdinand-engelbeen.be/klimaat/klim_img/dco2_em4.jpg
        Where the red line is the above calculation.
        The calculated rate of change is midst of the noise, caused by fast responses of (tropical) vegetation on fast temperature and drought changes (Pinatubo, El Niño/La Niña)…
        Bart makes a lot of the “pause” in the CO2 rate of change in the atmosphere by plotting the rate of change of the emissions and the increase in the atmosphere with different scales and an offset, while my plot simply shows that the natural variability in sink rate is relative huge (halve the emissions for yearly averages) and there were more such flat periods, even negative trends, but the overall trend is about halve the emissions and still widely within the natural variability.
        The natural variability is about +/- 1 ppmv (yearly averages) around the trend. That is (near) entirely caused by temperature. The trend of over 70 ppmv since 1960 is only for ~4 ppmv caused by the temperature increase since 1960, the rest is from human emissions…
        Where Bart goes wrong is by assuming that both the variability and the trend are caused by the same process…

    • rgbatduke
      You say, “earn some math before you engage in a QUANTITATIVE discussion on math.”

      You may be mesmerized by fancy differential equations that are use to fit curves, but the simple math behind the mass balance argument shows that even with increaseing temperatures, mother nature is absorbing HALF of human emissions, while the other HALF goes to increasing atmospheric concentrations by roughly 2 ppm per year. One must learn to balance a check book before learning how to integrate a function.

  58. @Willis
    “All you’ve done by such comments is shown that not only are you not following the story, and not only do you truly not understand the processes used and the physics in play, but that you want to be an arrogant jerk about it”
    I am following the story in fact and have at the very least a good grasp at the ‘physics at play’. The reason for my admittedly not particularly cordial comment was the ‘crack-pot’ comment you made and, well, the general arrogant demeanor that just seems to permeate from you whenever you decide to make a comment of these forums. I rarely comment on these forums or even take the time to read here, but when I do, your comments always do seem to stick out as being rather adversarial, and well, arrogant, hence my remark. Jaworowski is a crackpot, you say Willis? You do indeed waste everyone’s time by indulging in such infantile personal abuse of him. It is a pity that you should have resorted to these ad hominem attacks instead of engaging with his arguments properly, which you have totally failed to refute on any factual or rational grounds, and instead of supporting your own arguments with relevant facts and reasons too.
    @Ferdinand
    “At these parts of the ice core, huge differences were found in multiple samples. Neftel rejected the high levels, as these were highly probable from the contamination. I should have rejected all measurements of these parts of the ice as unreliable”.
    The issue is that Neftel arbitrarily deemed measurements as high as 417ppmv as ‘contained’ even though the measurements were taken from the same 1g samples as the others he decided to accept. It was arbitrary and there was no explanation for it aside from he thought ‘they were contaminated’. It seems to me that Neftel started off with a belief and a conviction – that atmospheric CO2 has not exceeded 400ppmv – and sought to gather evidence selectively to prove that conviction. That is not the way of scientists whose interest is only in finding out the truth.
    “I don’t know of any underestimate of CO2 in snow near the surface (citation?), Etheridge e.a. measured snow and firn top down from the surface and didn’t find a difference near the surface”.
    Etheridge seems to crop us a lot. When analysing the surface-snow at the Pionerskaya and Vostok station Raynaud and Delmas measured CO2 concentrations of 160ppmv and 240ppmv respectively while the concentration in the atmosphere at the time was 310ppmv. These were direct measurements and they imply that the ice-core probably underestimates paleo-atmospheric CO2 by at least 20-50%. This 20-50% underestimation of CO2 in the surface-snow would imply that when the ice-core measures 280ppmv during interglacials the actual CO2 concentration in the atmosphere at the time could have been as high as 420ppmv. This is something that would appear to accord nicely with Stomata proxy which shows CO2 as high as 430ppmv during the Holocene according to Margret Steinthorsdottir et al 2013. I am not arguing that the ice-core is useless. It is useful for shedding light on the climate switching between glacial and interglacial states and the relationship between CO2 and temperature but it has too many issues in my mind for it to be thought of as faithfully representing absolute CO2 values thousands of years ago.
    “With the wet method, all ice is melted and the CO2 extracted with vacuum, but the acid-carbonate reaction goes on, thus increasing the CO2 release over time. That is the reason that the wet method is completely abandoned and CO2 levels from the Greenland ice cores are not reliable and not used”.
    Mmm. I fail to see why that would be an issue myself. The ice can be melted in a hermetically sealed container and the total DIC content can then be measured.
    “Which didn’t show any extremes”.
    417ppmv is ‘extreme’? Based on what? The adjusted ice-core data? Whenever I hear someone say that any CO2 above 400ppmv is extreme or unprecedented and point to the ice-core data I roll my eyes because just about every aspect of adjusted paleoclimate ice-core reconstruction seems extremely conjectural and uncertain to me. I agree with Bart and Ronald. The ice-core data has to many issues for us to base any concrete conclusions from it.

    • Richard January 29, 2015 at 8:45 am

      @Willis

      “All you’ve done by such comments is shown that not only are you not following the story, and not only do you truly not understand the processes used and the physics in play, but that you want to be an arrogant jerk about it”

      I am following the story in fact and have at the very least a good grasp at the ‘physics at play’. The reason for my admittedly not particularly cordial comment was the ‘crack-pot’ comment you made and, well, the general arrogant demeanor that just seems to permeate from you whenever you decide to make a comment of these forums. I rarely comment on these forums or even take the time to read here, but when I do, your comments always do seem to stick out as being rather adversarial, and well, arrogant, hence my remark. Jaworowski is a crackpot, you say Willis? You do indeed waste everyone’s time by indulging in such infantile personal abuse of him. It is a pity that you should have resorted to these ad hominem attacks instead of engaging with his arguments properly, which you have totally failed to refute on any factual or rational grounds, and instead of supporting your own arguments with relevant facts and reasons too.

      Thanks for your comments, Richard. Is Jaworowski a “crackpot”? I already retracted that comment. But the problem is that his arguments don’t hold water. You say that I was not “engaging with his arguments properly, which [I] have totally failed to refute on any factual or rational grounds”.
      In fact, I’ve specifically pointed out and falsified several of his arguments, like his claim that there is no difference between ice age and air age … are you supporting him on that point?
      I also pointed out the error in his foolish claim that a 3% increase in an equilibrium situation is meaningless, and I gave an example and an explanation to back that up as well.
      I also pointed out that his claim that somehow the ice in the bubbles is lost or mixed with the current air by supplying a picture of intact bubbles in the ice core.
      Since you haven’t replied to any of those, I’d say that YOU are the person who is “not engaging with his arguments properly”. Which is why I said that you were “not following the story”.
      As to whether my comments are “adversarial”, in part science is an adversarial blood sport, wherein one person tries to falsify the cherished claims made by another person … how could that process be anything but adversarial? Do I get over the top sometimes … indeed. In the last two days I’ve shoveled eight tons of gravel, so I’ve likely been a little techy …
      Regards,
      w.

    • Richard,
      Are you sure that the measurements At Vostok were in snow? I only find mentions of much lower CO2 levels in ice, not snow by both researchers and others:
      http://www.nature.com/nature/journal/v284/n5752/abs/284155a0.html
      where the “too high levels in ice” were probably from Greenland ice cores, and Vostok was one of the first (if not the first?) Antarctic core…
      Further:
      http://www.nature.com/nature/journal/v315/n6017/abs/315309a0.html same story
      http://www.nature.com/nature/journal/v303/n5916/abs/303410a0.html
      Anyway, more recent measurements, not only by Etheridge, but also by others at Vostok and Dome C show CO2 in firn which are top down hardly below atmosphere…
      Some parts of the Siple Dome ice core were clearly contaminated with drilling fluid, reason why Neftel rejected the high values and choose the values which were in line with the samples before and after the contamination. But even so there was a large variability in his results. Later another core was drilled and didn’t show large variations at the same depths.
      Sorry Richard, stomata data are proxies, ice core records are smoothed direct measurements of CO2 in ancient air. If the average of the stomata data is different from the ice core data over the period of resolution of the ice core, then the stomata data are wrong.
      Why should one use a method that gives even more problems with the CO2 levels, if we have faster methods with a better accuracy? Carbonates in ice are measured too as part of dust deposits.
      I know, Bart and you don’t like ice core data…

      • I don’t like the ice core data because it is not subject to independent verification.
        But, it does not present any particular problem for the theory that atmospheric CO2 is driven by a temperature modulated process, which excludes human forcing from having any significant influence. dCO2/dt = k*(T – Teq) is what is happening today, and since at least 1958. There is no reason that it has to be the same for all time.

      • Bart, the ice core data are backed by stomata data and indirectly by coralline sponges for the period 1900-1980 and there is an overlap of ~20 years (1960-1980) with direct measurements.
        If the ice core data were far off, the stomata data would show huge abnormalities either up or down with a better resolution (but worse accuracy). The same for the coralline sponges: these measure δ13C in the ocean surface layer with a resolution of 2-4 years. δ13C in the ocean follows δ13C in the atmosphere with ~1 year equilibrium rate. If there was a huge CO2 release/uptake by vegetation or the oceans, that would show up in the δ13C record as a huge change up or down. Which doesn’t exist: only a smooth, increasingly faster drop in δ13C in ratio to human emissions…

      • 1900-1980 is a recent time interval. There is no guarantee that processes operating over geologic time do not corrupt the outcome.

      • Bart,
        Coralline sponges are highly accurate and high resolution measurements (not proxy’s) for δ13C in seawater, as the carbonate layer has exactly the same δ13C level as the surrounding seawater. Be it that this is rather uniform in the North Atlantic Gyre (Bermuda), but less uniform near upwelling places like the equatorial Pacific where sudden shifts are frequent. The Bermuda sponges give a good impression of the changes in the ocean surface and the atmosphere over the past 600 years, as both ocean surface and atmosphere are in close contact. For the period before 1850, there was little variation in δ13C level. After 1850, one can see an increasing drop in δ13C level in parallel with the increase of CO2 in the atmosphere and human emissions:
        http://www.ferdinand-engelbeen.be/klimaat/klim_img/sponges.gif
        Stomata data are proxy’s as they are growing over land, where a bias and huge variability of CO2 levels compared to “background” are frequent, besides other influences (droughts, nutrients,…) on the stomata (index).
        For the bias can be compensated by calibrating the stomata index data against direct measurements since 1960 and ice cores before 1960. While the stomata index data are calibrated against ice core data over the past century, these parallel each other. If the ice cores were far off, say, missing a drop or increase of 20 ppmv over a few years, that would be noticed in the higher resolution stomata data as a sudden jump in index and in sponges as a sudden jump in δ13C level.
        Stomata are less reliable further back in time, as the local bias can shift tremendously, depending of land use changes in the main wind direction, except if you calibrate them again with ice cores for earlier periods (and not point to stomata data as “proof” that the ice core data are wrong, as some here do)…

  59. if CO2 was really 30-50% lower in snow than in the surrounding air, I would expect the same problem further down where the snow is compacting to firn and ice.

    And shouldn’t you also expect the same problem further down where the fern is compacting to glacial ice.
    Does not the density of both the snowpack and the fern keep increasing from top to bottom (vertical axis) due to the force of gravity?
    And if the density is increasing from the bottom to the top of the vertical column ….. then the air (and CO2) containing “open spaces” within the column will be decreasing proportionally, …. and if so, shouldn’t the increase in pressure at the lower portion of the column be forcing the air (and CO2) up the vertical column via the lesser compacted “open spaces”?

    Snow crystals are extremely open: it is a little ice and a lot of air.

    Yes, snow crystals are extremely open prior to their accumulation into snowpack on the surface. But the snow crystals quickly get broken up and form into relatively round mature ice grains, as per stated or defined by, to wit:

    To understand the changes in the snow pack through time, we must first consider the mechanical breakdown of the intricate, delicate, crystals of falling snow into relatively round mature ice grains. The delicate arms of a snow crystal are easily broken. Constant motion and vibration of the water molecules breaks off the delicate exterior features. Ultimately, these pieces lodge in the spaces between the crystal’s arms. Through this process of destructive metamorphism, the original crystal eventually becomes a rounded ice ball.
    In nature, crystals lose their points due to molecular motion, wind, and direct pressure. Physically breaking the snow crystals, for instance stomping on them or disturbing them with a shovel, will produce the same effect. The crystal arms are broken and then rounded grains fuse by freezing into larger crystals in a process called sintering.
    http://ankn.uaf.edu/Curriculum/Athabascan/ObservingSnow/Snow.html

    Other than physical location …. is there a difference between “glacial snowpack” and “mountain forest snowpack”?
    If not, then to wit:
    Persistent wind-induced enhancement of diffusive CO2 fluxes in a mountain forest snowpack
    http://adsabs.harvard.edu/abs/2010AGUFM.B11D0393B

    • Samuel, that is an interesting reading, which shows that the decay of organics in winter simply goes on under a snow deck. The porosity of snow is high enough to allow the diffusion of CO2 to the atmosphere. Wind enhances the fluxes with up to 10%.
      As there is not much vegetation in Antarctica, there are no CO2 fluxes from vegetation decay and all what happens is that wind sheer and pressure pumps some air in and out, including the CO2 it contains. As far as I know that doesn’t make a difference in CO2 levels and several measurements in firn confirm a small change with depth, due to the increase in the atmosphere and the time needed for diffusion in smaller and smaller pores with depth…

  60. rgb,
    That was another very worthwhile comment. You have a knack for getting to the heart of the matter. So, thanks. Your comments add excellence to this site.
    Ferdinand Engelbeen and I argued a few years ago over his contention that human-emiitted CO2 was the cause of the recent rise. Ferdinand was extremely patient and polite. He never lost his temper, but instead, he used logic and facts.
    The result? Ferdinand convinced me. I was unable to refute his facts, so he dragged me [kicking and screaming] to the point where I was finally convinced, and I acknowledged his position. Now, that’s the way I think whenever I see related discussions: the past century’s rise in CO2 is due almost entirely to human emissions.
    I thanked Ferdinand at the time, and I’ll repeat that thanks now. His patience is a model, and his understanding of the issues is due to lots of study. He is certainly an asset here — as is Bart. Between them this will probably be all ironed out in due course, as more information becomes available.
    One more thing. While Ferdinand is very knowledgeable, he often neglects to take the final step: after all the arguing, and facts, and analysis, and discussion, too often he leaves off the conclusion: at current and projected concentrations, CO2 is harmless. And it is beneficial. More CO2 is clearly better for everyone — especially for the world’s poorest, who are hit hard by even small increases in the price of food.
    Since the biosphere measurably benefits from the rise in that harmless trace gas, I would hope that Ferdinand writes his conclusion of the matter more often, as he did a while back:

    “In my opinion even a doubling would have little impact, as clouds are a negative feedback (while all current GCM’s include clouds as a positive feedback!), thus a doubling of CO2 would have only moderate (and thus globally positive) effects.”

    So even a doubling of CO2 would be beneficial. There is no identifiable downside or negative to the rise in that healthy and harmless trace gas. More really is better, at least up to a doubling [which would anyway take a very long time].
    That conclusion needs to be stressed more often, because if it is accepted there is no need for demonizing “carbon” [other than the easy money that some people collect by scaring the public]. Rather, we should acknowledge that although the rise in CO2 was not really planned, we are very fortunate that it has turned out to be a good thing for the biosphere — and harmless as well.

    • One more thing. While Ferdinand is very knowledgeable, he often neglects to take the final step: after all the arguing, and facts, and analysis, and discussion, too often he leaves off the conclusion: at current and projected concentrations, CO2 is harmless. And it is beneficial. More CO2 is clearly better for everyone — especially for the world’s poorest, who are hit hard by even small increases in the price of food.

      That’s because it is important to divorce the science from the value judgements and assertions that one can solve a pair of coupled Navier-Stokes equations in one’s head! We do not know that additional CO_2 is, or will be in the future, harmless. It probably isn’t true. Some aspects of it may be beneficial. Some may not. Some of them may well lead to economic or human “catastrophes”. Others may be a great blessing. They can even be both — terrible for people living in coastal lowlands where even a half-meter sea level rise and wonderful for farmers in the great plains or central India. Cost-benefit analysis, risk analysis, is almost absurdly difficult because we don’t have models that work.
      Let me repeat that, very loudly:
      We Do Not Have Models That Work!
      That doesn’t mean that the future is safe. It means that we cannot compute the risk, or that our conditional computations of risk are subject to great uncertainty. The proper role of science is to be honest about this uncertainty so that people do not make critical decisions about how to invest scarce resources now based on claims of knowledge of future state that are made politically without any sound and defensible scientific or statistical basis.
      Don’t get me wrong, I can’t solve the Navier-Stokes equations in my head, either, and I don’t even know if I have a good grasp on all of the variables that are important enough to put into a crude model (like the ones that are being run — all enormously crude and incomplete and arbitrary) that might compute it 80 years into the future.
      That’s one of the things I like about Ferdinand’s contributions. He doesn’t usually state an opinion about whether human contributions to global warming are good or bad. He limits his participation mostly to defending the IMO very, very likely fact that surplus atmospheric CO_2 post 1850 almost entirely comes from human (and possibly occult geological) sources. If nothing else, there is a compelling correspondence. It may or may not be unique, it might or might not be true, but it is sensible, consistent, and to me rather plausible, given our limited knowledge of relaxation times and processes. Similarly, there is an entirely compelling correspondence between increasing CO_2 and increasing global average temperature, supported by a simple physical model that predicts an increase in temperature in good agreement with the observed increase in temperature, without any feedbacks. I am far from certain that this model will work without bound into the future, because again there is good reason from the geological record that at some point CO_2’s marginal effect is quenched by other feedbacks in the nonlinear and highly complex and chaotic system, but the fit between the simple physical model (one that ignores the fluid dynamics altogether) and data over the last 150 years is excellent.
      Hence my own belief: Humans have mostly caused, and continue to cause, an increase in atmospheric CO_2. Atmospheric CO_2 increases has caused, and continues to cause, a variation in the “dynamical equilibrium” temperature of the planet that is almost entirely logarithmic in CO_2 concentration (as predicted by the theory), where a total climate sensitivity including all feedbacks of around 1.8 C is in excellent agreement with the past. It is entirely plausible, but not certain, that is shift in equilibrium will continue into at least the immediate future as CO_2 continues to increase, provided that natural variations do not either enhance or cancel all or most of this change and provided that the system remains in the vicinity of the same attractor it is in now or isn’t perturbed in other ways by forces natural or anthropogenic. It is my personal — not scientific — opinion that the cost in lives, graft, and immediate human misery associated with premature and urgent demonization of coal-based electrical power is not justified by the amelioration of an almost incomputable future risk from this likely warming, given the high probability that long before the expensive measures being taken now have any impact besides perpetuating the poverty of the 1/3 of the world’s poorest people technological advances in the developed countries plus the increasing cost of carbon based fuels plus mere human greed and self-interest will cause a real reduction in the rate at which carbon is being burned.
      But do not resent the fact that many people in the scientific community, even those that are do not believe it particularly likely that future warming will be “catastrophic”, are unwilling to make a similar statement. For one thing, they may not understand the impact of the current policies on poverty. For another, they may not have ever visited countries with extreme poverty and seen what it really means. I grew up in India. I cannot imagine condemning the people of India, Africa, China, to decades and decades of life and death in 18th century conditions while enjoying 21st century benefits personally, just because there is a risk that carbon dioxide burned to provide them with cheap energy and better lives now will cause a further increase in global average temperatures of a degree or two over the next century, especially not with the prospect of e.g. thermonuclear fusion wiping out the carbon-based energy industry overnight, or photovoltaics plus things like LFTR wiping it out more slowly as coal gets ever more expensive and the alternative energy sources get ever cheaper and better supported by technology. If we do nothing at all I expect our utilization of carbon will peak long before the middle of the century.
      rgb

      • “Atmospheric CO_2 increases has caused, and continues to cause, a variation in the “dynamical equilibrium” temperature of the planet that is almost entirely logarithmic in CO_2 concentration (as predicted by the theory), where a total climate sensitivity including all feedbacks of around 1.8 C is in excellent agreement with the past.”
        It isn’t possible. CO2 evolves according to
        dCO2/dt = k*(T – Teq)
        This is a positive gain system between T and CO2. If, in turn, there is a significant positive feedback from CO2 to T, then that comprises a positive feedback loop with exponential instability.
        This is true even with the moderating influence of T^4 radiation. Result: the aggregate response of T to CO2 concentration, accounting for all ameliorating feedbacks, must be small.

      • Thanks RGB…
        It seems that we have quite the same ideas about CO2 and poverty: although grown up in Western Europe (in a far from rich, but happy family), I have been sailor (engine room) instead of (then obliged) military service and to have a (paid) view of the world, including the slums, the poverty (and the corruption) that I have seen in South America…
        Only one point: I fear that nuclear fusion still is a long way to go before getting commercial and that many of the intermittent alternatives lack the necessary huge storage to make them valuable. But one never knows…
        I once read a book in our (historical old) local library, with as title: “The Newest Human Inventions” or some similar title from…1901. Like a “dust sucking broom” (vacuum cleaner) and a lot of other devices which we now use like they where always there… Thus it may go much faster than I expect…
        BTW, If you meet Gary N. Greenberg somewhere at Duke (if he still is there) from the OEM-discussion list, please do greet hem from me…

      • Well, Bart, all I can do to counter your unsupported assertion that it is not “possible” (contingent on you being right on something I think you are wrong about) is present this graph:
        http://www.phy.duke.edu/~rgb/Toft-CO2-PDO.jpg
        I would say that CO_2 concentration plus a weak sinusoidal oscillation with a period of around 67 years is an almost perfect explanation for the mean temperature of the last 165 years. That doesn’t mean that it is correct, of course — in spite of the fact that the CO_2 dependent part is based on pure physics — but it means that your assertion that it is not possible is not, in fact, correct. It is possible. It is easily possible. It is a good fit.
        BTW, note well that — by your own argument, as you just brought up the weak oscillation yourself — it completely explains the “pause” in spite of the fact that temperature isn’t tracking CO_2 concentration at all. You simply have the situation backwards.
        On average, global temperature appears to be tracking CO_2 concentration as it increases. The relationship is just about exactly what one expects, that is to say, functionally related as the log of the CO_2 concentration. Superimposed on this is what appears to be a lovely sinusoid that I have no explanation for and which could be pure garbage, a chaotic epicycle that will vanish as unexplainedly as it appeared. Or, it could be a regular feature into the distant past that extends into the indefinite future. Lacking anything like good data into the more distant past, we are not likely to ever know.
        As for the CO_2 concentration increase itself. I’m curious about your explanation for the functional form that you are quite insistent must be right. Above in one place or another you assert that the cause for the temperature dependence on the rate of CO_2 increase is something like upwelling oceanic water. Fine, suppose that you are right. Mauna Loa clearly shows us that atmospheric CO_2 is increasing at the rate of ~2 ppm over the last decade or three. This corresponds – by simple arithmetic — to an extra 10 GT of atmospheric CO_2 per year (again in very round numbers). That is, I don’t care what somebody asserts is “the cause” or “the source”, the measurement from ML implies that this year there are roughly 10 GT more atmospheric CO_2 than there were last year.
        CO_2 in the atmosphere does not come with a label (ignoring the weak signature from isotopic analysis as too much trouble to work through in this very simple argument). A molecule of CO_2 could have come from recently burned coal or the churn of the biosphere or from deep upwelling ocean water and we cannot tell once it is mixed into the atmosphere.
        So I’m very puzzled. It is a simple matter of fact that humans dumped roughly 10 GT of CO_2 into the atmosphere last year. This CO_2 did not just “disappear”. It joined CO_2 being given off by the biosphere, CO_2 being absorbed by the biosphere, CO_2 being given off by the ocean, CO_2 being absorbed by the ocean. All of the inputs mix with it. It (plus the existing load plus the inputs) exit as it cycles around in the various reservoirs through the various processes. Yet you are certain that the increase is due to temperature according to your formula with its unexplained and meaningless (as far as I can tell) T_{eq}, and have boldly asserted further that all of the increase is coming from the oceans.
        How does that work out? If humans had not added 10 GT that just happens to correspond in quantity to the new 10 GT that appeared in the atmosphere, the system would have dropped by 10 GT in spite of the oceanic contribution? Or are you asserting that even if we had added 0 GT the CO_2 would have gone up by 10 GT? In the latter case, exactly how does the rest of the system “know” how to absorb human contributed CO_2 but not ocean-generated CO_2?
        rgb

      • “…all I can do to counter your unsupported assertion…”
        It’s very simple, RGB. The positive response from temperature to CO2, coupled with a significant positive response from CO2 to temperature, is unstable, even with T^4 radiation.
        Assume temperature reacts according to
        dT/dt = -a*T^4 + b*CO2
        for proportionality constants a and b. The perturbation system is
        ddT/dt = -(4*a*T0^3)*dT + b*CO2
        where dT is the perturbation of the temperature from any steady, stable value T0.
        Couple that with
        dCO2/dt = k*dT + k*(T0 – Teq)
        The characteristic equation for this system is
        lambda^2 + a*lambda -b*k = 0
        The roots are at
        lambda = -a/2 +/- sqrt((a/2)^2 + b*k)
        There if b and k are both positive, there is always one positive root, which means there is no stable equilibrium.
        It is possible to have a small value for b if, for example, we add a small additional term
        dCO2/dt = k*(T – Teq) – c*CO2
        for some value of c small enough that it is not observable over the years since 1958. But, that b is small. Sensitivity of temperature to CO2 simply cannot be significant, or the Earth’s climate would have reached a tipping point long ago.
        Temperature data from before 1880, at best, is not at all reliable (and, so what if it is? This isn’t necessarily a stationary system). Since at least 1880, temperatures are well modeled as a trend, plus ~60 year cyclicality, the pattern laid in well before CO2 could have caused it.
        “…in spite of the fact that the CO_2 dependent part is based on pure physics…”
        No. It is only pure physics when taken in isolation of feedbacks and other potentially confounding processes. The statement, “All things being equal, increased atmospheric CO2 will increase surface temperatures”, is correct. The “all things being equal” part is the rub.
        ” It joined CO_2 being given off by the biosphere, CO_2 being absorbed by the biosphere, CO_2 being given off by the ocean, CO_2 being absorbed by the ocean.”
        There’s your clue. The amount given off by the biosphere is orders of magnitude larger. Whatever processes deal with that input can easily deal with our puny imputs.
        “Yet you are certain that the increase is due to temperature according to your formula with its unexplained and meaningless (as far as I can tell) T_{eq}, and have boldly asserted further that all of the increase is coming from the oceans.”
        It is due to a temperature modulated process. There is no doubt about it. Human inputs are not temperature modulated.
        I have not asserted that it is all coming from the oceans. I have said that is a reasonably likely source.
        “In the latter case, exactly how does the rest of the system “know” how to absorb human contributed CO_2 but not ocean-generated CO_2?”
        Because it’s a feedback system. It reacts to the total volume. It takes in both. But, the natural sources are open ended – the ocean/atmospheric interface, for instance, compels a certain atmospheric concentration, and cannot be resisted – while human inputs are finite, and relatively small.
        Ferdinand – the total gain isn’t less than unity in this system. See above.

      • I was in a hurry. Should have been:
        The characteristic equation for this system is
        lambda^2 + (4*a*T0^3)*lambda -b*k = 0
        The roots are at
        lambda = -(4*a*T0^3)/2 +/- sqrt(((4*a*T0^3)/2)^2 + b*k)
        Or, just substitute a for 4*a*T0^3. The point is, there is no way to stabilize this system in a manner consistent with observations for any significant positive value for b.

  61. OK, let me see if I can bring a bit of order to the discussion. We have three methods proposed to estimate the modern changes in CO2. One is Bart’s method, which uses the difference in temperature from some baseline value to estimate the change in CO2 (∆CO2, or d(CO2)/dt). The second is from Ferdinand, which uses the total cumulative emissions to estimate atmospheric CO2. Finally, we have the exponential decay method, which assumes that the “extra” CO2 will decay at some rate given by an “e-folding time” to some pre-industrial “equilibrium” value.
    Here’s the difficulty:
    https://wattsupwiththat.files.wordpress.com/2015/01/co2-estimated-by-three-methods.jpg
    As you can see, we simply don’t have enough data, or perhaps enough variety in the data, to determine which one is correct.
    In terms of the mean squared error, they are all quite close. The exponential decay (e-folding) is the best, the cumulative emissions is second, and the ∆CO2 is third.
    My own vote goes to the exponential decay, for four reasons:
    1. It has the best fit (smallest squared error).
    2. Fitting the parameters of the exponential equation using e-folding time and pre-industrial CO2 value as the fitted parameters, gives a fitted pre-industrial value of 283 ppmv, which agrees well with the ice core values and other estimates.
    3. Exponential decay is a very common phenomenon in nature.
    4. Bart’s method has two problems. First, it implies a very large sensitivity of CO2 to temperature. But there is a larger problem.
    Bart’s equation (which I used to create the good fit above) is
    dCO2/dt = k*(T – Teq)
    The problem is that if the temperature T becomes constant at some value greater than Teq, the equation says that CO2 will continue to increase forever … which makes no sense at all.
    I hope this cuts through some of the misunderstandings … but then this is climate science …
    w.

    • “The problem is that if the temperature T becomes constant at some value greater than Teq, the equation says that CO2 will continue to increase forever … which makes no sense at all.”
      As I have admonished before, it is a mistake to extrapolate a local model far beyond the bounds of its verified region of applicability.
      I gave the example earlier of a transistor. A transistor provides a gain which is the local slope of the input/output relationship at a particular bias voltage. If you extrapolate that linearized behavior to zero bias voltage, you would conclude that a transistor amplifier would amplify inputs with no input power supply, which is of course nonsense.
      There is no doubt that a temperature modulated phenomenon has been driving atmospheric CO2 in the modern era – see above. How long that phenomenon preceded 1958, and how long it will persist into the future, is anyone’s guess. However, the inertia of the past will probably keep it going for a while yet.
      I will give you one plausible scenario in which the phenomenon could have arisen. There are surely others. Only further investigation can reveal the actual mechanism.
      Suppose that, during the pre-industrial past when CO2 levels appeared to be quiescent, surface waters of the oceans were in equilibrium in pCO2 between the oceans and atmosphere. Everything stayed nice and steady in the atmosphere, as dictated by the atmospheric/ocean interface.
      Now, suppose that a portion of the vast pools of liquid CO2 reported to linger in the depths burped to the surface, creating a long term steady flow of CO2 enriched waters to the surface. At each incremental time step, the surface waters ineluctably become a little more enriched with CO2, forcing the atmosphere to continuously equilibrate. Over time, there is a steady increase in pCO2 of the surface oceans, and the interface forces a corresponding steady increase in atmospheric CO2.
      The rate at which the ocean CO2 outgasses to the atmosphere is a function of temperature. This results, for as long as the flow from the depths lasts, in an atmospheric increase which obeys
      dCO2/dt = f(T)
      where f(T) is the temperature dependent flow rate into the atmosphere. This can be expanded in a first order Taylor series as
      f(T) = k*(T – Teq)
      for some values k and Teq to be determined by observation. The forcing only exists for as long as the flow from the burp persists.

    • BTW:
      “As you can see, we simply don’t have enough data, or perhaps enough variety in the data, to determine which one is correct.”
      Variety, indeed. That is why it is worthwhile to plot the rate of change, which has much greater variety visible to the naked eye. Do that, and it will quickly become apparent which model matches the data best.

      • Thanks, Bart. I agree that the temperature influences the monthly or annual CO2 levels, through several mechanisms. So it is no surprise when we look at ∆CO2, temperature has a large effect. And in fact, the correlation of ∆T and ∆CO2 is 0.36.
        The problem is that to that relationship, your formula adds an inexorable increase in CO2, regardless of whether temperature stops increasing or not. This gives the false match to inexorable increase in CO2 in the historical data.
        The main thermal mechanism relating ∆CO2 and ∆temperature that I see is the outgassing of CO2 from the ocean as it gets warmer. However, both Henry’s Law and ice core data agree as to the size of that increase in CO2.
        As a result, that mechanism is far too small to explain the ~125 ppmv increase in CO2 that we’ve seen over the last century and a half.
        Next, your formula makes no logical sense. Your method says that if the temperature drops, the CO2 keeps rising. I’ve gotten a good fit, but your formula says that if the temperature reversed course and started dropping tomorrow exactly as it has risen, and in 35 years it retraced its steps and returned to where it was in 1979, CO2 would continue to go up over the entire period. Note that you claim that your formula is valid over that entire temperature range, so there is no “extrapolation” of any kind …
        Why would that be? What physical process would work that way, that a thirty-five year drop in temperature would cause a constant increase in CO2 every year? Your formula is non-physical, it makes no sense at all.
        Finally, oceanic outgassing doesn’t explain the change in the isotopic carbon makeup of the atmosphere.
        Best regards,
        w.

      • Willis – it is not a given that the natural world is stationary and quiescent. Please read the part above where I start with “I will give you one plausible scenario in which the phenomenon could have arisen.” You should see that your objections are not applicable there.

      • @ Bart January 29, 2015 at 2:43 pm

        There is no doubt that a temperature modulated phenomenon has been driving atmospheric CO2 in the modern era – see above. How long that phenomenon preceded 1958, and how long it will persist into the future, is anyone’s guess. However, the inertia of the past will probably keep it going for a while yet.

        Beautiful statement, Bart, simply beautiful ….. and “right as rain”.
        Bart, me thinks your naysayers can’t keep their minds focused on the “temperature” in question.
        When you are talking about the past, present or future temperature or average temperature of the ocean waters …… their minds immediatelly associate your stated temperature to the “fuzzy” math calculated Average Global/Regional near-Surface air Temperatures.
        And that is equivalent to associating or comparing ….. apples to rutabeggers, … and very little of what you say makes sense to them when they do it.
        As far as I am concerned, the only logical explanation for the “steady & consistent” average 6 ppm bi-yearly cycling of atmospheric CO2 … and the average 2 ppm yearly increase in atmospheric CO2 that has been recorded over the past 57 years … is a direct result of the seasonal temperature change in ocean’s surface waters …. and the gradual yearly increase in the average temperature of said ocean waters.
        Just like most any deep lake water could be warming up even though its surface water is still freezing and thawing as the seasons change.

      • Samuel,
        Both the seasonal variations and the 2-3 years responses on Pinatubo and El Niño are reactions of vegetation on temperature variations.
        Seasonal:
        http://www.ferdinand-engelbeen.be/klimaat/klim_img/seasonal_CO2_d13C_MLO_BRW.jpg
        Short term:
        http://www.ferdinand-engelbeen.be/klimaat/klim_img/temp_dco2_d13C_mlo.jpg
        In both cases, the CO2 changes and δ13C are opposite to each other and vegetation is the dominant factor. If the oceans were the driving force, CO2 changes and δ13C changes would parallel each other.
        But the overall trend is not caused by vegetation: vegetation is a growing sink for CO2. Thus short term variation and longer term trend are from different processes, where the trend may be caused by the oceans or human emissions or temperature…
        Temperature change is by far not enough to cause such a CO2 change and the oceans have the wrong (positive) δ13C level to be the cause of the current CO2 increase and δ13C drop…

      • Both the seasonal variations and the 2-3 years responses on Pinatubo and El Niño are reactions of vegetation on temperature variations.

        Gimme a break, Ferdinand, the 92’ Pinatubo and 98’El Niño events had no real effect on US corn production …… and so goes the corn, so goes the rest of the vegetation, and plenty of pretty graphs for you to see herein, to wit:
        http://www.farmdoc.illinois.edu/marketing/mobr/mobr_08-02/Tables%20and%20Figures%20for%20MOBR,%20Feb%202008.pdf
        And Ferdinand, here is the actual Mauna Loa data that you apparently massaged or squeezed to get what you were looking for out of, to wit:
        Maximum CO2 ppm per calendar year 1990-2013
        1990 5 1990.375 357.29
        1991 5 1991.375 359.09 +1.80
        1992 5 1992.375 359.55 +0.46 Pinatubo
        1993 5 1993.375 360.19 +0.64
        1994 5 1994.375 361.68 +1.49
        1995 5 1995.375 363.77 +2.09
        1996 5 1996.375 365.16 +1.39
        1997 5 1997.375 366.69 +1.53
        1998 5 1998.375 369.49 +2.80 El Niño
        1999 4 1999.292 370.96 +1.47
        2000 4 2000.292 371.82 +0.86
        2001 5 2001.375 373.82 +2.00
        2002 5 2002.375 375.65 +1.83
        2003 5 2003.375 378.50 +2.85
        2004 5 2004.375 380.63 +2.13
        2005 5 2005.375 382.47 +1.84
        2006 5 2006.375 384.98 +2.51
        2007 5 2007.375 386.58 +1.60
        2008 5 2008.375 388.50
        2009 5 2009.375 390.19
        2010 5 2010.375 393.04
        2011 5 2011.375 394.21
        2012 5 2012.375 396.78
        2013 5 2013.375 399.76 +2.98
        Ferdinand, please note in the above that the “Maximum CO2 ppm” occurred in the month of May (5) from 1990 up to and including 1998, the year of sid El Nino, which produced, generated or caused a 2.80 ppm increase in atmospheric CO2.
        But what caused the 2.51 ppm increase in 2006 and the 2.98 ppm increase in 2013?
        And why is your plotted CO2 data and temperature data pretty much completely out of synch with one another on your above graph?
        And Ferdinand, note the “Maximum CO2 ppm” occurred in the month of April in the two years (1999, 2000) following the 1998 El Nino. So, explain how your Average Surface Temperatures caused that to occur. Did the corn start sucking up the CO2 a month earlier in 1999 and 2000?

        Temperature change is by far not enough to cause such a CO2 change and the oceans have the wrong (positive) δ13C level to be the cause of the current CO2 increase and δ13C drop…t

        Ferdinand, show me a graph of your “daily maximum δ13C” measurements …. and/or cite a link to the “raw data” for said.

      • Samuel,
        Have a better look at the graphs: there is an opposite variability between CO2 changes and δ13C changes: that is caused by vegetation, not the oceans. Not in your corn, but the much larger amount of carbon in the NH forests (seasonal) and the tropical forests (Pinatubo, El Niño).
        Have a look at the daily change in δ13C, measured by Keeling Sr. at Big Sur state park (California) in the early 1950’s, which shows the negative correlation between δ13C and CO2 changes over a day:
        http://www.ferdinand-engelbeen.be/klimaat/klim_img/diurnal.jpg
        More can be found at:
        http://www.biokurs.de/treibhaus/literatur/keeling/Keeling_1955.doc
        For more recent data in the bulk of the atmosphere:
        http://www.esrl.noaa.gov/gmd/dv/iadv/
        And as you know CO2 levels always follows temperature changes with a ~90 deg. lag, as a sudden increase in temperature gives a sudden change in CO2 fluxes, but that doesn’t give a sudden increase in CO2 level: that needs time to build up… Taking the derivatives will shift both derivatives 90 deg. back in time, thus still a 90 deg. lag of dCO2/dt after dT/dt.

      • there is an opposite variability between CO2 changes and δ13C changes: that is caused by vegetation, …

        Ferdinand, read this circa 2000 abstract. Not the entire document, just down to the “Introduction” and then get back to me on your “caused by vegetation” thingy, …. to wit:
        http://link.springer.com/article/10.1007/s004420000609#page-1
        And be sure to remember the part about ….. “discrimination decreased with altitude in stemwood”.

        δ13C changes: that is caused by vegetation, not the oceans. Not in your corn, but the much larger amount of carbon in the NH forests (seasonal) and the tropical forests (Pinatubo, El Niño).

        Really now, Ferdinand, do you have proof of that or are you just guessing?
        Here are your NH forests, with by far the greatest majority residing in the far northern latitudes and consisting of almost 100% conifers. Also pictured is your tropical forests wherein the Sunlight, temperature and/or precipitation does not vary significantly enough during a yearly or multi-year cycle to adversely affect their growth cycle.
        Of course the bi-yearly “flooding” cycle in some locales of tropical forests will have a tremendous effect on the outgassing of CO2 as the result of the rotting and decaying of the accumulated dead biomass …. which is why the soil in tropical forests are extremely poor in nutrients. To wit, Global Forests and Woodlands:
        http://www.technologystudent.com/joints/forest1a.png

        Have a look at the daily change in δ13C, measured by Keeling Sr. at Big Sur state park (California) in the early 1950’s,

        Ferdinand, do you know where Big Sur State Park is located? Here, take a look-see, to wit:
        https://over60hiker.files.wordpress.com/2014/03/am-map-of-big-sur.gif
        A coastal location with high humidity ocean breezes a blowing in and out on a daily basis.
        And Ferdinand, as far as the “early 1950’s” is concerned, best you read about something a closer to your home that you can possibly relate to, ….. to wit:

        A Scandinavian group accordingly set up a network of 15 measuring stations in their countries. Their only finding, however, was a high noise level. Their measurements apparently fluctuated from day to day as different air masses passed through, with differences between stations as high as a factor of two.
        Charles David (Dave) Keeling held a different view. As he pursued local measurements of the gas in California, he saw that it might be possible to hunt down and remove the sources of noise. Taking advantage of that, however, would require many costly and …..
        http://www.aip.org/history/climate/co2.htm

      • Samuel,
        I have not the slightest idea of what you are trying to prove with that. The δ13C levels in conifers may be influenced by drought and sunlight, so what? Does that influence the day/night δ13C and CO2 changes and to what extent? Does that influence the seasonal δ13C and CO2 changes and to what extent? Further there is also a lot of (dwarf) birch wood up to the polar seas, which shed their leaves in fall and grow new ones in spring.
        The largest changes in diurnal CO2 (and δ13C) are found in forests when there is no wind. The lowest changes if there is incoming wind form the oceans. Again a proof that the oceans are a lot slower than vegetation for seasonal CO2 and δ13C changes and hardly any diurnal CO2 changes.
        Further, drought is an additional factor in the 2-3 years changes caused by an El Niño: large parts of the tropical forests are drying out and while the decay of plant debris goes on for a few years (until the “fuel” is reduced), CO2 uptake is reduced or release even increased by wildfires.

  62. Bart January 29, 2015 at 2:43 pm

    “The problem is that if the temperature T becomes constant at some value greater than Teq, the equation says that CO2 will continue to increase forever … which makes no sense at all.”

    As I have admonished before, it is a mistake to extrapolate a local model far beyond the bounds of its verified region of applicability.
    I gave the example earlier of a transistor. A transistor provides a gain which is the local slope of the input/output relationship at a particular bias voltage. If you extrapolate that linearized behavior to zero bias voltage, you would conclude that a transistor amplifier would amplify inputs with no input power supply, which is of course nonsense.

    Sorry, but that’s not a parallel situation. You claim that the “region of applicability” includes the temperature T. But if the temperature T stays constant, even for a decade, the CO2 continues to rise. That makes no sense, and there’s no “extrapolation” at all.
    w.

    • It’s an example of extrapolating a linearized approximation of a nonlinear function beyond the region of validity.
      Read the scenario above and see if you still find it implausible.

      • If you mean the scenario of the non-existent “vast pools of CO2”, I couldn’t make sense of it.
        w.

      • I thought I read somewhere about liquefied CO2 at the ocean floor. Can’t find the article anymore.
        Doesn’t matter. It could be a bubble of CO2 upwelling which formed from past downwelling, and is now rising to the surface. Transport phenomena are quite complex.
        If you find a way to shoot down a mechanism I am suggesting, don’t just stop there. Use your imagination and come up with a better scenario. We know… without a doubt, we know… that atmospheric CO2 is evolving according to
        dCO2/dt = k*(T – Teq)
        This can arise from transport of extra CO2 into the surface of the oceans, or from other sources. The objection that “your formula adds an inexorable increase in CO2, regardless of whether temperature stops increasing or not” is facile and not the knock-out punch you seem to think it is.

      • Bart, you don’t seem to realize that your equation
        dCO2/dt = k*(T – Teq)
        is simply a linear regression of the form
        y = mx + b
        As such, given that x and y have some statistical relationship, you’ll be able to find an “m” and a “b” that give you a reasonable fit … but that fit does NOT mean that there is physical causation going on.
        One way that we can tell if there might be such a physical causation is to explore what happens under various scenarios. For example, suppose variable “x” starts decreasing instead of increasing. Or suppose that variable “x” becomes constant. What does the given model tell us in such cases?
        If in such cases the relationship still holds, then you MIGHT have a valid causational relationship. On the other hand, if your relationship stops working in such cases and you need to re-fit the equation to give you a new “m” and a new “b”, then you can be pretty sure that you’re just engaged in curve-fitting, and that there is no such underlying causational relationship. No guarantees, but at that point the odds that you’re looking at causation are really, really bad.
        Now, note that in the case of both Ferdinand’s relationship (airborne C = 0.57 * emitted C) or the exponential decay model (e-folding time = 58 years, pre-industrial value = 283 ppmv), we don’t need to recalculate our parameters. Both models give physically possible outcomes under either steady emissions or decreasing emissions with no change in the parameters, although the outcome of the exponential decay model seems much more likely to me.
        But in the case of your model, it falls apart completely as soon as the temperature either goes steady or starts decreasing.
        All the best,
        w.

      • Willis,
        I agree that the CO2 decay calculation gives a much better fit, the 0.53*emissions is just a nice first approximation and only coincidence as the emissions increase slightly quadratic over time, which gives a near linear increase in rate of change. My factor of 0.53 is somewhat lower than yours, as I didn’t take into account land use changes, as these are quite uncertain. The factor 0.53 therefore is only for direct burning of fossil fuels and cement manufacturing…
        For Bart’s change in factors to fit other periods, see the graph further down for the period 1900-1960 with the same parameters as for fitting 1960-2011…

      • Willis Eschenbach @ January 29, 2015 at 8:37 pm
        “Bart, you don’t seem to realize that your equation
        dCO2/dt = k*(T – Teq)
        is simply a linear regression of the form
        y = mx + b”

        Yes, it is that. But, it is much more than that.
        There is not simply agreement between the slope and offset, which would be trivial. There is agreement across every frequency component evident.
        The odds of that being happenstance are virtually nil.

      • “I agree that the CO2 decay calculation gives a much better fit.”
        A superficial fit in a domain where there is very little variability to match.
        Do the fit in the rate domain, where you can see all the bumps and burbles. There is no question in that domain which provides the better fit, and it is
        dCO2/dt = k*(T – Teq)

      • Yes, and that is what I have been saying.
        The second is inferred from the phasing of the data, and the trend in temperature matching the trend in the rate of change of CO2, along with all the other activity in dCO2/dt and temperature.

    • Willis,
      The formula of Bart indeed shows a constant increase of CO2 with a sustained change in temperature.
      He doesn’t take into account that the increase of CO2 in the atmosphere gives an increase in pressure which pushes more CO2 in the oceans (and plant alveoli), which makes that any temperature change gives a finite increase of CO2 in the atmosphere of ~8 ppmv/K (historical) and less on short term: ~5 ppmv seasonal and 4-5 ppmv short term (2-3 years).
      Bart’s formula need another k and Teq for about all other periods on earth (for each period another X ppmv/K/year), while the 8 ppmv/K is valid for all periods over the past 800,000 years (except the past 160 years, but still valid for the variability).

      • No, Ferdinand. Continuous upwelling of CO2 enriched waters produces an ineluctable increase in atmospheric CO2.
        You keep thinking in static terms. This is a dynamic system.

      • No Bart, any increase of CO2 from upwelling waters (either by increased concentration or increased deep ocean upwelling) is met with an increase of pCO2 of the atmosphere, which increases the uptake by and decreases the input from the oceans. To give a continuous increase of CO2 in the atmosphere, the upwelling must increase a 4-fold in the time span 1959-2011, which is what human emissions did. Here the reaction of the fluxes plotted for a 10% increase in upwelling followed by a 1 K increase in temperature:
        http://www.ferdinand-engelbeen.be/klimaat/klim_img/upwelling_incr_temp.jpg
        There is a small quadratic term if there was an increase in temperature together with more upwelling, but most should come from a 4-fold increase of the upwelling itself.
        There is evidence that the ocean upwelling increased: not in local ocean pCO2 measurements at the upwelling places, not in the δ13C sink rate, not in the 14C/12C sink rate after the nuclear bomb tests spike, not in the residence time estimates…

      • If the continuous input of fresh extra CO2 in the surface oceans were incapable of driving atmospheric CO2 as you suggest, then human emissions could not drive it either.
        In both cases, it is a direct injection of fresh CO2 into the surface system. You can’t have it both ways. You cannot treat fresh injections of CO2 differently depending on source.

      • Bart, that is a matter of decay rate: the current sinks show a pressure (difference) related e-fold decay rate of ~51 years, by far not fast enough to influence the short term changes caused by temperature swings, either seasonal or short term (2-3 years) and they show a large part of human emissions or any sustained natural source large enough to be measurable in the atmosphere.
        If there was a 10% increase in CO2 concentration of the deep ocean upwelling, that would lead to an initial increase of ~20% in CO2 influx, due to the large increase of pCO2 at high equatorial temperatures.
        With a relative slow decay rate, that builds up in the atmosphere, but the increased pressure in the atmosphere reduces the influx (which is directly proportional to the pCO2 difference between atmosphere and ocean surface) and increases the outflux (again proportional to the pCO2 difference).
        That gives that with an increase of ~30 ppmv CO2 in the atmosphere, everything is in equilibrium again at a higher throughput: from the current ~40 GtC/year to ~45 GtC/year:
        http://www.ferdinand-engelbeen.be/klimaat/klim_img/upwelling_incr.jpg
        Al what an additional increase in temperature does is pushing the new equilibrium between ocean inputs and outputs some 8 ppmv/K higher (4-17 ppmv/K in the literature, 8 ppmv/K historical).
        In all cases a new equilibrium is reached after a few decades for a one-step increase in upwelling or temperature or human emissions. Only if there was a year after year increase in upwelling or temperature or human emissions, the CO2 levels in the atmosphere would go up in ratio with the emissions (natural or not) or temperature.

      • No, Ferdinand. This is a continuous flow of high pCO2 waters. It continually pushes the pCO2 of the surface ocean, and hence of the atmosphere, higher.
        There is no difference between this and human emissions, except that it can be orders of magnitude greater. Both are continuous inputs of fresh CO2 to the surface system. Both can potentially drive atmospheric concentration higher. But, only one can be large enough to explain the observed rise if sinks are aggressive.

      • Bart, there is already a continuous flux of CO2 into the atmosphere: app. 40 GtC/year from the deep oceans at the equatorial upwelling places which continuously returns into the deep oceans mainly via the THC sink place in the NE Atlantic.
        A constant CO2 influx doesn’t change anything in the atmosphere as long as that is balanced by a constant sink rate. Which is what happens after some time when the CO2 pressure in the atmosphere and in the oceans equilibrate. That is a dynamic equilibrium with a lot of exchange but no changes in CO2 levels.
        If for some reason one of the inputs shows a positive step change in flux rate, all what happens is that the CO2 pressure in the atmosphere increases to a new higher level where influx and outflux again are in equilibrium. Whatever caused the step change (upwelling, temperature, humans), the atmospheric pressure adjusts to a new equilibrium: that is a negative feedback to any disturbance of the ocean – atmosphere CO2 equilibrium.
        That simply means that whatever the initial change in influx caused by temperature, upwelling or humans, the change in pressure in the atmosphere will suppress the change (Le Châtelier’s principle) by adjusting the in and out fluxes, until a new equilibrium is reached at a different CO2 level in the atmosphere.
        You can’t have a fixed extra CO2 flux into the atmosphere from the oceans without a reaction of the change in CO2 pressure on the in/out fluxes of the same oceans…
        Only a continuous increase like the recent human emissions or a continuous increase in temperature like during glacial-interglacial transitions will give a continuous change in CO2 level.

      • Yes, Ferdinand, a continuous increase in surface ocean pCO2 will cause a continuous increase in atmospheric CO2. That is what I have been saying.
        Upwelling of CO2 enriched waters diffuses into the surface waters over time, causing a continuous increase in pCO2.

      • Bart that is only possible if there is a continuous increase in temperature and/or upwelling, not for a one-step increase in temperature (as your formula says) or a one-step increase in upwelling. The first has stalled, the second has not the slightest indication in any observation…

      • Yes, and that is what I have been saying.
        The second is inferred from the phasing of the data, and the trend in temperature matching the trend in the rate of change of CO2, along with all the other activity in dCO2/dt and temperature.

    • “You claim that the “region of applicability” includes the temperature T. But if the temperature T stays constant, even for a decade, the CO2 continues to rise.”
      And, a time interval.
      BTW, temperatures have been staying relatively constant for the past decade, and CO2 has been rising linearly, in line with the differential equation.
      Emissions are accelerating, i.e., the total running accumulation of emissions would be rising at least quadratically. That is a rather severe disconnect in the human attribution scenario.

      • Bart January 29, 2015 at 5:16 pm

        BTW, temperatures have been staying relatively constant for the past decade, and CO2 has been rising linearly, in line with the differential equation.
        Emissions are accelerating, i.e., the total running accumulation of emissions would be rising at least quadratically. That is a rather severe disconnect in the human attribution scenario.

        You don’t seem to have understood either Ferdinand’s graph or my graph above. Both of them show that the relationship between cumulative emissions and CO2 is very near to LINEAR, including for the past decade.
        This means that your claim that CO2 is rising linearly and cumulative emissions are rising quadratically is … well … demonstrably incorrect. If that were true, you wouldn’t get the good linear fit between the two.
        And your method does no better than Ferdinand’s method at estimating CO2, in fact of the three methods shown yours is the WORST.
        In any case, the exponential decay model is the one that matches the data best. It is also supported by the fact that such relationships are quite common in nature.
        w.

      • Willis – you lose a lot of definition in these plots of accumulated quantities. You cannot see the divergence unless you focus in on the specific area.
        The information contained in the derivative is the same as that in the overall quantity, modulo an integration constant. Look at the numerical derivatives. By focusing in on these plots of absolute totals, you are blinding yourself to the important detail.
        There is not doubt at all, in the rate domain, that my model fits better than the others. Plot it for yourself and see.
        Ferdinand –
        A) the ice cores may or may not be valid
        B) There is no prohibition against a sudden change in the factors for the linearized model. There is no reason to conclude that the model must be stationary. There is no reason one might not require a gain schedule.

      • Bart,
        You are overfocused on the details by looking at the derivative: all short term variations are from the influence of short term temperature variations on vegetation, but the trend is NOT caused by vegetation. Thus the very good (for a natural process) correlation between temperature variation and CO2 rate of change says next to nothing about the cause of the trend, which is from a different process.
        Willis’ and my plots focus on the trends and ignore the small variability around the trend caused by the temperature variability. After all, the whole temperature increase of ~0.8°C since the depth of the LIA may be good for ~6 ppmv CO2 increase in the atmosphere (according to Henry’s law), while we are looking at the cause of an increase of 110 ppmv over the equilibrium for the current temperature.
        There is one known process which increased slightly quadratic over time: human emissions. These fit the trend as well as in absolute levels as in the derivative. There may be natural processes which increased also a fourfold in the period 1959-2011 synchronous with the human emissions, but until now I haven’t seen any evidence for such an increasing natural input.

      • You have to focus on the details, Ferdinand. They are like the fingerprints left at the crime scene. If you cannot fit the fine detail, then you do not have a working model, and you cannot determine the culprit for the rise.

      • Said another way, a model which matches both the fine detail and the coarser detail is a better model than one that only matches the coarse detail. Matching the coarse detail is really not very challenging. Matching the fine detail is what tells us we have the correct model.

      • Bart,
        All what the focus on the details show is that there are small (+/- 1 ppmv from one year to the next) CO2 variations around the trend, which is currently ~110 ppmv above the equilibrium for the current ocean temperature. Or for the period 1959-current 70 ppmv extra.
        The short term processes are certainly temperature driven. The increase itself may be temperature driven or may be influx driven, with or without any influence from temperature. You can’t know the cause of the increase from looking at the temperature variations, as variation and trend are certainly caused by different processes…
        As noticed further on: while your temperature only factor fits the current period modestly well (either amplitude or trend, not both), you need to constantly change the parameters to fit other periods in the past, while our decay rate process fits all periods, from ice ages to interglacials and back, the MWP-LIA transition and the current increase in the atmosphere…

      • Doesn’t matter how small you think it is. If I can explain it, and you cannot, then my model is better than yours.

      • Bart,
        My explanation is as good (or better) than yours:
        – (near) all variability around the trend is caused by temperature variability by its influence on (tropical) vegetation, as the opposite CO2 and δ13C variations show.
        – (near) all of the trend is caused by human emissions, as these are about twice the increase in the atmosphere and neither oceans or vegetation can be responsible for the increase.
        – there is not the slightest connection between the variability around the trend and the cause of the trend…

  63. Willis wrote at 4:14 pm:

    “The main thermal mechanism relating ∆CO2 and ∆temperature that I see is the outgassing of CO2 from the ocean as it gets warmer. However, both Henry’s Law and ice core data agree as to the size of that increase in CO2.”

    The basic problem is see is too many people treat the oceans as if they were sterile bodies of salt water, warming and cooling through the millennia. The oceans are bio-geochemical living entities. Of course, not in any sensient manner, but the oceans have had 3.5 billion years of teeming microbial lifeto establish complex webs of feedback and homeostasis. They have had chlorophyll-based life for at least the past 1.7 Gy. This is part of the emergent properties hypothesis of Earth’s climate homeostasis.
    The phytoplankton contribution has been discussed before on WUWT.
    This NASA’s take on it:
    here: http://earthobservatory.nasa.gov/Features/Phytoplankton/
    http://earthobservatory.nasa.gov/Features/Phytoplankton/images/diatoms_CO2.png
    I would submit that once we have several years of OCO-2 data, this picture will become much clearer where the CO2 sinks and sources are and their seasonal magnitudes.
    My money is on OCO-2 data doing to the anthropogenic CO-2 Climate Change meme what the 16th-17th century telescope data did for 1,400 years of Ptolemaic conceptions of geocentrism.

    • Thanks, Joel. I would submit that your graph is the output of a computer model which has never been tested … and upon further research, I find it’s worse than that. The paper says:

      Using a global model of ocean biogeochemistry coupled to a climate model, we explore the effect of climate change on the distribution of diatoms, a key phytoplankton functional group. Our model results suggest that climate change leads to more nutrient-depleted conditions in the surface ocean and that it favors small phytoplankton at the expense of diatoms. At 4xCO2, diatoms relative abundance is reduced by more than 10% at the global scale and by up to 60% in the North Atlantic and in the subantarctic Pacific. This simulated change in the ecosystem structure impacts oceanic carbon uptake by reducing the efficiency of the biological pump, thus contributing to the positive feedback between climate change and the ocean carbon cycle. However, our model simulations do not identify this biological mechanism as a first-order process in the response of ocean carbon uptake to climate change.

      If someone buys that, I’ve got a bridge to sell them … it’s models all the way down.
      w.

  64. Thanks Willis for coming back to this thread.
    I am just getting back to this thread discussion after 24 hrs, since it is so far down now on WUWT.
    I agree with the model comments from Willis. It’s models all the way down for NASA. But at first principle, increases in SST could lead to less CO2 uptake by phytoplankton, hence rising atmospheric pCO2. Which will reverse once the SST variability starts to go down and phytoplankton uptake of CO2 goes up. This would conceptually be the phase lag that Bart discusses and is of biological in origin.
    My BIG recommendation is for Bart, Ferdinand, and Willis come to a “consensus” on the discussions above, centered around Bart’s equations using the T data-sets as test cases, and then to collaborate on a joint paper in AW’s new Atmo journal. THE OCO-2 level 2 data will be released around March 2015, so the southern ocean CO2 levels and SST will make for interesting test cases in the coming years.

  65. Joel O’Bryan January 30, 2015 at 10:45 pm

    But at first principle, increases in SST could lead to less CO2 uptake by phytoplankton, hence rising atmospheric pCO2. Which will reverse once the SST variability starts to go down and phytoplankton uptake of CO2 goes up. This would conceptually be the phase lag that Bart discusses and is of biological in origin.

    In general, increasing temperature (within limits) increases the growth of plants. We see this on land as winter turns to summer. As a result, I see no theoretical reason that an increase in SST would lead to LESS CO2 uptake by plankton.
    Next, I see no reason to figure there will be a “reverse” in the CO2 uptake once “SST variability starts to do down”. In fact, I don’t even understand what that means. Why would the SST variability decrease?
    So I’m sorry, Joel, but none of that make sense.

    My BIG recommendation is for Bart, Ferdinand, and Willis come to a “consensus” on the discussions above, centered around Bart’s equations using the T data-sets as test cases …

    Bart has already admitted that his equation only works in selected circumstances and for certain times. It doesn’t work if the temperature goes stable. It doesn’t work if the temperature goes down. It doesn’t work during the ice ages. It only works for a short given period of time.
    This is because Bart’s equation is merely a bozo-simple linear regression of the temperature and the dCO2, of the form
    y = m * x + b
    In this case the function is:
    dCO2/dt = m * T + b
    If you think that his equation represents a physical causative phenomenon, think again. It’s just curve fitting.
    The part that Bart doesn’t seem to get is that there is a difference between the short-term relationship between CO2 and temperature, and the long-term trend. He seems mightily impressed by the short-term correlation between CO2 and temperature caused by the outgassing and uptake of CO2 from the oceans as they heat and cool.
    And that is indeed happening, just as Bart claims. But it does NOT explain the ~120 ppmv increase in CO2 that we’ve seen over the last century and a half.
    So I see no hope of either Ferdinand or I ever coming to a “consensus” with Bart on this question. If his equation were valid, it would work at all times and whether temperature was going up, down, or stable. It doesn’t. It’s not valid. Yes, in the short term temperature and CO2 are inter-related … but that’s not the explanation for the trend.
    The part I don’t understand is why people are so insistent on denying that humans putting 9 gigatonnes of carbon into the atmosphere every year is NOT the reason that atmospheric CO2 is increasing. What’s the problem with accepting that? Here’s the carbon cycle. It’d a decade or so old, and human emissions have gone up from the 5.5 Gt/yr since then:
    https://wattsupwiththat.files.wordpress.com/2012/05/carbon_cycle_diagram.jpg
    Now … why would the additional carbon from burning fossil fuels NOT increase the atmospheric CO2?
    And again, why (as both Ferdinand and I have shown) is there such a close relationship between human emissions and the increase in CO2?
    And again, why (as Ferdinand showed with the corals) has the ∂13C in the atmosphere changed in exactly the direction we’d expect from the burning of fossil fuels?
    With all of the problems with the “CO2 roolz” theory, I don’t get why folks get stuck on this question. I wrote a post about that, you might take a look.
    Anyhow, don’t hold your breath for either Ferdinand or I to agree with Bart. Not gonna happen. He’s just doing simplistic curve fitting, and as soon as his magic formula doesn’t work, he just re-fits his formula with new parameters and claims success … sorry, but the world doesn’t work like that.
    w.

    • The part I don’t understand is why people are so insistent on denying that humans putting 9 gigatonnes of carbon into the atmosphere every year is NOT the reason that atmospheric CO2 is increasing.

      You’ve got me, there, too. Especially when Mauna Loa shows an increase of (guess what) roughly 9 GT of carbon dioxide in the atmosphere every year. This leads to the question (which Bart studiously avoided answering above) “What do you think the atmospheric CO_2 concentration would be if we added zero GT to the atmosphere instead of the roughly 10 corresponding to the observed increase?”
      I suppose one could stubbornly insist that the answer would still be 9 GT higher, but that really is a silly response unless one thinks that the equilibration time for CO_2 is weeks or months, not decades. But if it were weeks or months it wouldn’t be anywhere near as smooth a curve as it is.
      I also couldn’t make any sense at all out of his assertion that CO_2 doesn’t have a local equilibrium concentration or his avoidance (again) of the observation that his equation for the derivative of CO_2 concentration over time is unstable even at a constant temperature, as long as that temperature is above “T_eq”. Whatever T_eq means.
      From my own numerical experiments with the CO_2 concentration, it is really pretty easy to create a model that fits it post 1957 based on all sorts of assumptions, simply because since Mauna Loa started taking data it has been almost perfectly monotonic and boring and anybody can fit a nearly linear trend. Some of the models might make sense, some might not. Even Bart’s isn’t necessarily senseless if one assumes that the ocean itself is heavily disequilibrated in such a way that it outgasses more CO_2 every year than it re-absorbs. The fundamental question isn’t whether or not one can build a model that empirically works well enough to model the data that might be justified if that assumption is true, it is whether or not that assumption is true. It seems to me that direct measurements of CO_2 dissolved in the ocean — the so-called “acidification” of the ocean — are if anything evidence of the opposite, that the ocean is acting as a net sink for the atmosphere. However, there is probably room in our ignorance of the thermohaline circulation and global state of the ocean to allow the possibility without necessarily considering it the best, most likely possibility.
      And in the end, no matter how you slice it, there is a very strong correspondence between anthropogenic CO_2 additions and the increase in atmospheric concentration.
      rgb

      • “You’ve got me, there, too. Especially when Mauna Loa shows an increase of (guess what) roughly 9 GT of carbon dioxide in the atmosphere every year.”
        MLO data shows an increase of roughly half of human emissions.

  66. Willis:
    You say/ask

    The part I don’t understand is why people are so insistent on denying that humans putting 9 gigatonnes of carbon into the atmosphere every year is NOT the reason that atmospheric CO2 is increasing. What’s the problem with accepting that?

    There is a difference between
    (a) denying that humans putting 9 gigatonnes of carbon into the atmosphere every year is NOT the reason that atmospheric CO2 is increasing
    and
    (b) questioning if humans putting 9 gigatonnes of carbon into the atmosphere every year CAN BE the reason that atmospheric CO2 is increasing.
    You are conflating (a) and (b).
    A scientist assesses the available data and seeks additional data to determine if the small (relative to natural) human emission of 9 gigatonnes of carbon is or is not a significant contribution to the increase in atmospheric CO2. A scientist does NOT “accept” either possibility without good reason.
    At present, the evidence of the dynamics of the seasonal variation in atmospheric CO2 suggests the human emission of 9 gigatonnes of carbon is NOT a significant contribution to the increase in atmospheric CO2, but available data does not confirm or refute this suggestion.
    One can hope the accumulation of data from the OCO2 satellite will resolve the matter.
    Richard

    • Richard,
      The net global seasonal change is ~5 ppmv, with a ~2 ppmv increase every year.
      After three years, the trend is larger than the seasonal “noise”.
      Further:
      – the seasonal changes are proven from (NH extra-tropical) vegetation
      – the 2-3 years variability in rate of change is proven from (tropical) vegetation
      – the trend is proven NOT from vegetation and NOT from the oceans.
      What else do you think is fast and large enough to give an increase in the atmosphere, completely synchronized with human emissions?

      • Ferdinand
        Please check down the thread. I am not avoiding your question but I cannot answer it until tomorrow because I have been given a ‘time out’ for withstanding bullying.
        Richard

      • I always enjoy it when someone accuses me or someone else on the web of “bullying” as Richard Courtney has done. It’s a joke because to bully someone, you have to be able to threaten them with something. That’s what a “bully” is, someone who uses threats to get their way, like the schoolyard bully saying “Give me your lunch money or I’ll beat you up.” To be a bully, you have to try to force someone to do something by using threats.
        But in a scientific discussion on the web, what is the threat? That I’m going to speak mean things about someone? That I’ll say bad words about their science?
        I cannot do anyone any harm, nor do I wish to harm anyone, nor do I have any way to harm them even if I did wish that. In fact, an “internet bully” in a scientific discussion is a chimera, an imaginary creature.
        For example, Greg Laden is trying desperately to get Willie Soon fired. So he’s threatening someone off the web, in real life. And even then he’s not a “bully”, because there’s no quid pro quo. There’s nothing he’s trying to force Willie to do. Laden is just being malicious. As a result, I’d say that Laden is a gormless stoneless underhanded bastard for trying to cost a man a job because he doesn’t like Willie’s scientific conclusions … but even Greg Laden is not an internet bully.
        So my rule of thumb is, if someone accuses me of being an internet bully, it’s a sure mark that I’ve won the debate.
        w.
        PS—And no, asking someone to take a time out until they can stay within the blog comment guidelines is not bullying either. It’s an attempt to maintain order.

      • Ferdinand:
        Sorry for the delay. This is my reply which attempts to provide a brief reprise of the issues which you and I have repeatedly disputed notably on WUWT and also mebtions why the OCO2 data may resolve matters.
        There are three main categories of opinion as to the cause of the recent rise in atmospheric CO2 concentration. I itemise each of them.
        1. The anthropogenic overload hypothesis.
        Many people (e.g. you, the IPCC, Robert Brown, Willis Eschenbach) observe the annual emission of CO2 from human activities (i.e. the anthropogenic CO2 emission) is greater than the rise in atmospheric CO2 concentration and say this suggests the anthropogenic emission is overloading the system such as to cause the rise. For convenience I am calling this the anthropogenic overload hypothesis.
        The three main problems with this hypothesis are:
        (a) the anthropogenic CO2 emission is only ~2% of the total (i.e. natural and anthropogenic) emission each year and few natural processes are significantly overloaded by such a small change as 2%. This is especially important because the anthropogenic emission is locally emitted and natural local emissions vary by more than this. This local variation is clearly demonstrated by the preliminary OCO2 data for atmospheric CO2 concentration plotted here and here where the anthropogenic emission from e.g. Europe is not discernible.
        (b) In some years, almost all of the anthropogenic CO2 emission seems to be absorbed into the sinks for CO2, and in other years almost none seems to be absorbed. The overloading should exist or not and not be intermittent. Supporters of the hypothesis smooth the data to obtain an apparent fit between the anthropogenic emission and the rise in atmospheric CO2 concentration.
        (c) Importantly, the observed dynamics of the system are not as would be expected by overloading of the sinks for CO2. This is the data for atmospheric CO2 concentration as recorded at Mauna Loa. The annual rise is the residual of the seasonal variation each year. The seasonal variation has a clear ‘saw tooth’ form: it rises rapidly and almost linearly then falls rapidly and almost linearly. There is no reduced rate of fall as the sinks for CO2 fill and, clearly, they do not fill prior to the rapid transition to seasonal rise in atmospheric CO2. Other CO2 measurement sites show the same effect. If the sinks don’t fill then the anthropogenic overload hypothesis is falsified.
        The only rational opposition to this conclusion I have seen was provided by Robert Brown in a WUWT post. I have failed to find that post so cannot link to it and will explain my understanding of the opposition anticipating that Robert Brown will correct me if my understanding is wrong. He points out that electrical circuits can provide similar ‘saw tooth’ signals by use of components such as capacitors which can switch from being electrical absorbers and emitters (as some mechanisms in the carbon cycle switch from being absorbers to emitters with the seasons). Hence, an analogue computer model of the carbon cycle could probably emulate the Mauna Loa data. Perhaps so, but it is hard to equate such switching with the anthropogenic overload hypothesis.
        2. The disturbed equilibrium hypothesis.
        Some people (e.g. me. Arthur Rorsch) think the carbon cycle is constantly moving towards an equilibrium which it never achieves and something has altered the equilibrium state with the result of the observed rise in atmospheric CO2 concentration.
        In this hypothesis, the cause of the rise is the cause of the changed equilibrium to the carbon cycle. The human emissions of CO2 may have caused the changed equilibrium or e.g. temperature rise in past decades may have changed the equilibrium. Some processes of the carbon cycle system are very slow with rate constants of years and decades. Hence, the system takes decades to fully adjust to the new equilibrium.
        And the observed dynamics of the carbon cycle fit with this.
        The seasonal variation in atmospheric CO2 is provided by effects of processes with rapid (i.e. hours, days, weeks, months) rate contents responding to the changing seasons
        while
        the observed rise in atmospheric CO2 is provided by effects of processes with long (i.e. years, decades, centuries) rate contents responding to altered temperature or anthropogenic emission or etc..
        Using this hypothesis, the rise in atmospheric CO2 concentration can be modelled as having an entirely natural cause, an entirely anthropogenic cause, or some combination of natural and anthropogenic causes.
        (ref. Rorsch A, Courtney RS & Thoenes D, ‘The Interaction of Climate Change and the Carbon Dioxide Cycle’ E&E v16no2 (2005) )
        Each of the six models in this paper matches the available empirical data without use of any ‘fiddle-factor’ such as the ‘5-year smoothing’ the IPCC uses to get its model to agree with the empirical data.
        So, if one of the six models of this paper is adopted then there is a 5:1 probability that the choice is wrong. And other models are probably also possible. And the six models each give a different indication of future atmospheric CO2 concentration for the same future anthropogenic emission of carbon dioxide.
        Data that fits all the possible causes is not evidence for the true cause. Data that only fits the true cause would be evidence of the true cause. But the above findings demonstrate that there is no data that only fits either an anthropogenic or a natural cause of the recent rise in atmospheric CO2 concentration. Hence, the only factual statements that can be made on the true cause of the recent rise in atmospheric CO2 concentration are
        (i) the recent rise in atmospheric CO2 concentration may have an anthropogenic cause, or a natural cause, or some combination of anthropogenic and natural causes,
        but
        (ii) there is no evidence that the recent rise in atmospheric CO2 concentration has a mostly anthropogenic cause or a mostly natural cause.
        3. The temperature response hypothesis.
        Other people (e.g. Bart) say the rise in atmospheric CO2 concentration can be explained as being a result of temperature change. And the clear relationship of CO2 and temperature over short times is said to support this hypothesis.
        It can be hoped that he OCO2 data will indicate the evolution of atmospheric CO2 concentration across the planet with the seasons. If so, then it may be capable of resolving which – if any – of the three above hypotheses is correct.
        Richard

    • Willis Eschenbach
      I am glad I gave you enjoyment.
      The record is what it is and your B…S… does not change it.
      I refuse to debate the issue with you because there is no value in obtaining more of your bullying which – when it fails – results in you running away while shouting insults over your shoulder.
      Richard

  67. richardscourtney January 31, 2015 at 12:27 am

    At present, the evidence of the dynamics of the seasonal variation in atmospheric CO2 suggests the human emission of 9 gigatonnes of carbon is NOT a significant contribution to the increase in atmospheric CO2, but available data does not confirm or refute this suggestion.

    Richard, do you really think that such handwaving without presenting the alleged “evidence” means anything? Because on my planet, such a claim is meaningless. Bring up, cite, and identify the evidence you so lightly refer to, or don’t bother with the waving of the hands …
    In addition, how come you are ignoring the evidence of the change in the ∂13C concentration in the atmosphere?
    w.

    • Willis:
      You reply to my pointing out your conflation of two different statements by saying

      Richard, do you really think that such handwaving without presenting the alleged “evidence” means anything?

      Say what!?
      Your evasion means much on this planet.
      I have repeatedly stated the evidence on WUWT. Indeed, you claim to be supporting the arguments of Ferdinand Engelbeen and he and I have been disputing these matters for more than a decade in many places including on WUWT . Indeed, you have disputed the evidence with me on WUWT. “Alleged”!?
      Any who are interested in the evidence can see e.g. this WUWT comment and consider all the discussion in that thread.
      As I said to you

      There is a difference between
      (a) denying that humans putting 9 gigatonnes of carbon into the atmosphere every year is NOT the reason that atmospheric CO2 is increasing
      and
      (b) questioning if humans putting 9 gigatonnes of carbon into the atmosphere every year CAN BE the reason that atmospheric CO2 is increasing.
      You are conflating (a) and (b).
      A scientist assesses the available data and seeks additional data to determine if the small (relative to natural) human emission of 9 gigatonnes of carbon is or is not a significant contribution to the increase in atmospheric CO2. A scientist does NOT “accept” either possibility without good reason.

      That is NOT “arm waving” because your assertion that the human emission must be “accepted” as the cause is NOT “good reason” and your evasion of my point suggests you know it is not “good reason”.
      Richard

      • Richard, I asked for something specific and simple—the evidence you mentioned in your claim that “the evidence of the dynamics of the seasonal variation in atmospheric CO2 suggests the human emission of 9 gigatonnes of carbon is NOT a significant contribution to the increase in atmospheric CO2, but available data does not confirm or refute this suggestion.”
        In response to this simple request, you’ve loaded up with accusations that I’m “evading” and “conflating” and who knows what … but you haven’t presented the evidence.
        As to conflating, your claim was that:

        There is a difference between
        (a) denying that humans putting 9 gigatonnes of carbon into the atmosphere every year is NOT the reason that atmospheric CO2 is increasing
        and
        (b) questioning if humans putting 9 gigatonnes of carbon into the atmosphere every year CAN BE the reason that atmospheric CO2 is increasing.
        You are conflating (a) and (b).

        Many folks on this very thread are asserting (a), and that’s what I was talking about when I said

        The part I don’t understand is why people are so insistent on denying that humans putting 9 gigatonnes of carbon into the atmosphere every year is NOT the reason that atmospheric CO2 is increasing. What’s the problem with accepting that?

        Now, indeed there are other folks out there who are “questioning if humans putting 9 gigatonnes of carbon into the atmosphere every year CAN BE the reason that atmospheric CO2 is increasing” … so what? That question to me is meaningless. Of course it CAN be the reason. Whether it is the reason or not is another question, but whether you like it or not, it is a possible explanation.
        So I wasn’t conflating anything, which is why I roundly ignored your claim that I was. I don’t have time to answer every foolish claim made on WUWT.
        w.

      • “The part I don’t understand is why people are so insistent on denying that humans putting 9 gigatonnes of carbon into the atmosphere every year is NOT the reason that atmospheric CO2 is increasing. What’s the problem with accepting that? “
        Because the data are contrary to it.
        Look, as far as AGW is concerned, it is only one piece of the puzzle. Why should I care if it is true or not as far as debunking AGW is concerned? I’m not.
        I’m concerned with it because the data show, unequivocally, that human inputs are not the driving factor in atmospheric CO2 concentration.

      • Willis
        I am replying to your additional evasion.
        You say

        Richard, I asked for something specific and simple—the evidence you mentioned in your claim that “the evidence of the dynamics of the seasonal variation in atmospheric CO2 suggests the human emission of 9 gigatonnes of carbon is NOT a significant contribution to the increase in atmospheric CO2, but available data does not confirm or refute this suggestion.”

        Yes, you did. And you asked it as a method to evade my point that you had conflated two issues which I stated and later quoted when pointing out that you had evaded my point.
        You now try to pretend that I raised your conflation as a method to avoid your question! NO!! You posed your question as a method to evade my clear – and iterated – point.
        Also, I gave you a link to the information you requested. If you don’t understand it then please ask for clarification, but don’t pretend I did not provide it especially when you and I have discussed the matter on WUWT in the past.
        This conversation is not productive. You made an assertion and I tried to get you to discuss your assertion; that conversation may be productive. I repeat

        your assertion that the human emission must be “accepted” as the cause is NOT “good reason” and your evasion of my point suggests you know it is not “good reason”.

        Importantly, the question you posed to evade discussion of your assertion is not important because, as I said,

        At present, the evidence of the dynamics of the seasonal variation in atmospheric CO2 suggests the human emission of 9 gigatonnes of carbon is NOT a significant contribution to the increase in atmospheric CO2, but available data does not confirm or refute this suggestion.
        One can hope the accumulation of data from the OCO2 satellite will resolve the matter.

        Sometimes available data is not capable of providing an answer, and data from the OCO2 satellite may resolve this matter.
        Richard

      • richardscourtney January 31, 2015 at 12:16 pm

        Willis
        I am replying to your additional evasion.
        You say

        Richard, I asked for something specific and simple—the evidence you mentioned in your claim that “the evidence of the dynamics of the seasonal variation in atmospheric CO2 suggests the human emission of 9 gigatonnes of carbon is NOT a significant contribution to the increase in atmospheric CO2, but available data does not confirm or refute this suggestion.”

        Yes, you did. And you asked it as a method to evade my point that you had conflated two issues which I stated and later quoted when pointing out that you had evaded my point.
        You now try to pretend that I raised your conflation as a method to avoid your question! NO!! You posed your question as a method to evade my clear – and iterated – point.
        Also, I gave you a link to the information you requested. If you don’t understand it then please ask for clarification, but don’t pretend I did not provide it especially when you and I have discussed the matter on WUWT in the past.

        I’m sorry, Richard, but what you gave me was a link to you claiming that smoothing was wrong, viz:

        But there can be no valid reason to smooth over 4, 5 or more years.
        The IPCC uses 5-year smoothing because using smoothing over shorter times fails to get agreement between the anthropogenic emissions and the rise in atmospheric CO2.

        I didn’t see anything in your link that I would consider to be evidence. No links to scientific studies. No mathematical analysis. No thing falsifiable, just more of your personal opinions … which is why I said you hadn’t provided any evidence.
        Anyhow, obviously this is going nowhere. I generally just ignore your comments, and I’ll go back to doing that. Answering them just leads to you abusing me and accusing me of “evasion” and “conflation” and “pretending”, and that’s no fun.
        w.

  68. Willis
    PS
    I don’t “ignore” the isotope ratio changes or any other evidence (as those who have followed my debates with Ferdinand will attest) but I see your raising that as a ‘red herring’ to avoid discussion of your ‘argument by assertion’.
    Richard

    • Willis
      Whatever excuses you use to run away from discussion of your assertion, it is clear that you have no ability to support your assertion that your opinion should be “accepted”.
      By chance your post concludes saying to me

      I generally just ignore your comments, and I’ll go back to doing that. Answering them just leads to you abusing me and accusing me of “evasion” and “conflation” and “pretending”, and that’s no fun

      and is followed in the thread by my post which explained why I was not answering your accusation that I was “ignoring” the isotope ratio changes.
      Pot, kettle.
      .And you suggest my illustrative link only addressed smoothing. No! Here it is again. It discusses many things and includes this

      I introduced the question by saying
      “A.
      The dynamics of the seasonal variation indicate the natural sequestration processes of the carbon cycle can easily absorb all – both natural and anthropogenic – emissions of CO2 in each year
      but
      B.
      The rise in atmospheric CO2 concentration measured at Mauna Loa since 1958 indicates the natural sequestration processes of the carbon cycle do not absorb all the emissions of CO2 in each year.”
      This matters because
      the residual of the variation in each year is the rise in atmospheric CO2 concentration of each year. If the natural sequestration processes of the carbon cycle did absorb all the emissions of CO2 in each year then there would be no rise.

      You made an assertion you cannot justify and your bluster does not disguise that.
      Richard

      • Richard, I’m reminding you once again that you are getting a bit out of line with your comments. This is not your domain. Take my advice: a 24 hour time out and cool off. Anthony

  69. “how come you are ignoring the evidence of the change in the ∂13C concentration”
    I think the change in atmospheric δ13C isn’t the bomb-shell evidence that the IPCC et al would have us believe. I have studied it and found that its argument is a very weak one which does not contradict or refute any of theories that the increase could be natural. The isotope signature of anthropogenic CO2 is very fuzzy due to being mixed up with the similar isotope signatures of CO2 emissions from other sources and the δ13C content from vegetation is almost the same from human fossil fuels. This means that, as Roy Spencer has pointed out, an increase in biomass decay could decrease the relative amount of C13 in the atmosphere and leave the same signature. To add to that could be bacteriogenic methane activity which is deficient in C13 and rapidly oxides into CO2 (especially under UV-light) and underwater hydrothermal vents. Whatever is causing the change in ∂13C however the idea that human CO2 is the cause is unlikely to my mind because the residence time for an individual atmospheric CO2 molecule (be it human or natural) is only around 4 years (3.8 years to be precise when using the IPCC’s figures in AR4). The isotope ‘evidence’ cannot be used as proof that the rise in CO2 is from fossil fuels because 1) a change in ∂13C is not a unique signature of anthropogenic CO2 and 2) the vast majority of anthropogenic CO2 has been absorbed by sinks and has become homogeneously intermixed with the pre-existing natural CO2 in sinks. This has already been confirmed by isotopic measurements. The atmospheric CO2 mass is in equilibrium with HCO3 (bicarbonate) and CaCO3 (carbonate calcium) at a δ13C permil value of -7 and anthropogenic (including biogenic) CO2 when in equilibrium has a δ13C permil value of -26. Assuming that humans increased the atmospheric CO2 content from 280ppmv in 1850 to approximately 351ppmv (about 25%) in 1988 this gives us a permil value of -11.75 (i.e. 25% of -26 and 75% of -7). However in 1988 Keeling measured a permil value of -7.807 implying that the maximum amount of anthropogenic CO2 residing in the atmosphere (including biogenic CO2) was about 4%, while 96% was entirely natural.
    “Are you sure that the measurements At Vostok were in snow?”
    That was the wrong paper. It was Raynaud and Delmas 1977. They measured that the surface-snow underestimated atmospheric CO2 by 20-50%. So it is not that Jaworowski is making an argument himself that the ice-core may underestimate atmospheric CO2 by 50%, these are based on measurements of the surface-snow. These are direct measurements and should not be disregarded so quickly.
    “Some parts of the Siple Dome ice core were clearly contaminated with drilling fluid”
    They were “clearly” contaminated? You state things as though they are matters-of-fact as if you know them without a shadow of doubt. Is that really so? Why were they clearly contaminated? Jawowoski explains the reasons as to why Neftel deleted the higher measurements. Quote: “A criterion for such data selection was the high CO2 concentration, and its discussion by Neftel et al. (1982, 1988) is rather fuzzy. They stated that ‘in larger samples (300g) contamination [with drilling fluid] is almost inevitable and the measured CO2 concentrations tend to be higher than the air originally included in the ice’. However, they did not measure CO2 in the air bubbles and secondary cavities from larger samples, but only from 1g samples. These small samples were used to ‘avoid samples with visible cracks’. They stated that, in the case of small (1g) samples, their ‘small size meant (they were) probably uncontaminated and we conclude that the lowest CO2 values best represent the CO2 concentrations of the originally trapped air’. But they recorded both the high and the low CO2 concentrations in 1g samples, so their conclusion is neither based on analytical evidence nor logical arguments”. Doesn’t sound they were clearly contaminated to me.
    “Stomata data are proxies, ice core records are smoothed direct measurements of CO2 in ancient air”.
    Sorry Ferdinand, but the ice-core is a proxy and just as with all proxies cannot provide strong evidence for anything at the end of the day simply because they are prehistoric and relate to times for which no direct empirical measures of anything are possible. Everything is conjectural with them. The ice-core cannot be accounted for in any empirical way and empirical measurements are impossible without a two-way time machine. In the absence of real evidence you can assert any theory you like, because others cannot show that you are wrong. But that does not make such theories right. Moreover, as others have pointed out, the ice-core is subject to many fractionation processes, such as gravitational compression which forces CO2 up to the surface, gases having different solubility coefficients, the formation of clathrates, and even chemolithotrophs which feed off CO2 and convert it into other chemicals, the list goes on. I will say though, that I don’t believe the increase in atmospheric CO2 is caused by the warming oceans from a reduction in the solubility of CO2 as others have assumed here. I think you calculated a figure of 1C/16ppmv whereas I have calculated a figure of 1C/19ppmv. I think the main point that skeptics of the idea that the increased in CO2 is largely anthropogenic here should be focusing on is the Revelle Factor, since that is what largely gives anthropogenic CO2 its very long lifetime, and as Segalstad has shown, the Revelle Factor is flawed and in violation of the fast-equilibria of Henry’s law and the 1:50 partitioning ratio. And yes, I am aware of your counterargument to this Ferdinand and, with all respect, I personally consider it hooey.

    • Richard
      Thankyou. I have made the same points concerning the isotope ratios in many places including on WUWT. Indeed, when I was pointing out the discrepancy in the 1990s I had to get people to ‘do the sums’ for themselves before they would agree that the isotope data proves nothing.
      But such matters are not directly pertinent to the consideration of ice core data.
      Ice cores and stomata each provide useful proxy data for past atmospheric CO2 concentration. Their usefulness differs and neither provides a direct indication of past atmospheric CO2 concentration.
      Richard

    • “The ice-core cannot be accounted for in any empirical way and empirical measurements are impossible without a two-way time machine. In the absence of real evidence you can assert any theory you like, because others cannot show that you are wrong. But that does not make such theories right. “
      Well stated.

    • Richard,
      You need to take into account all evidence together, not only one by one at a time
      There are only two sources of low δ13C on earth: recent organics and fossil organics, both much lower than the atmosphere. All other carbon sources are higher in δ13C than the atmosphere. That is the (deep) oceans, carbonate rock, volcanic vents/eruptions, etc…
      That effectively excludes (deep) ocean emissions or any other known source as the cause of the increase of CO2 in the atmosphere, except the biosphere and fossil fuel use.
      Thus while the timing and ratio of the δ13C decrease in the atmosphere is clearly linked to the use of fossil fuels, it is not possible to make a differentiation between fossil fuels use and vegetation decay on the base of the δ13C levels.
      But there is a way out: the oxygen use. Each type of fossil fuel uses its own amount of oxygen. Taken all together, slightly less oxygen is used than calculated from fossil fuel use. Thus the whole biosphere (plants, microbes, molds, insects, animals) is a net producer of oxygen (the earth is greening) and thus a net absorber of CO2 (~1 GtC/year) and preferential of 12CO2, thus not the cause of the decline of the δ13C levels in the atmosphere. The sole cause is human emissions…
      Of course, methane is another (very) low δ13C source and increased from 0.7 to about 1.8 ppmv in recent times. But even if that was all natural (it is all human…), that is about 10% of the decrease, 90% still human CO2.
      Then the 30% decrease measured. That is quite simple: all human CO2 is injected in the atmosphere but spreads over the other reservoirs, which are higher in δ13C, as there are huge seasonal exchanges with the atmosphere. The decline of δ13C therefore is measured in the atmosphere, leaves and the ocean surface (e.g. coralline sponges). That is the first reduction of the influence of the human low-13C emissions. The most important reduction is by the deep oceans: what goes into the deep is the current isotopic composition of the atmosphere (minus some isotope shifts). What comes out of the deep is the composition of the deep oceans (at least 1000 years older than what goes in and largely mixed). That makes it possible to estimate the deep ocean – atmosphere – deep ocean fluxes:
      http://www.ferdinand-engelbeen.be/klimaat/klim_img/deep_ocean_air_zero.jpg
      With 40 GtC/year continuous CO2 exchange between ocean upwelling and sink places, the whole “discrepancy” between human emissions and the observed increase in the atmosphere is gone.
      The 40 GtC/year is also the estimate of deep ocean – atmosphere CO2 exchanges based on the 14C decay rate from the 14C nuclear bomb tests in the 1950’s.
      Richard, the residence time has nothing to do with the decay time of an excess amount of CO2 above equilibrium. The residence time is:
      mass / throughput = 800 GtC / 150 GtC/year = ~5.3 years
      the e-decay rate of an injection of extra CO2 in the atmosphere is:
      Excess / sink flux = 110 ppmv / 2.15 ppmv/year = ~51 years
      If humans should stop all emissions today, the 110 ppmv would be 55 ppmv after ~40 years, 27.5 ppmv after ~80 years,… While the residence time hardly would change.
      That was the wrong paper. It was Raynaud and Delmas 1977.
      Sorry, if you cite such important points, please give a direct link…
      All what I could find were lower levels in ice and a lot of links to not so skeptic sites (if it is for being skeptic for points they like) all pointing to… Jaworowski, not the most reliable person about CO2 in ice cores…
      And this gem from 1977 (my bold):
      Raynaud, D.; Delmas, R.
      IAHS-AISH Publication (1977), 118(Isot. Impurities Snow Ice), 377-81 CODEN: PIHSD9; ISSN: 0144-7815. French.
      The composition of gas, extracted from polar ice samples, was determined It is close to the atmospheric composition with a higher CO2 and Ar content and a lower O2 content. Possible reasons for these differences are discussed.

      Are you really sure that the 30-50% less CO2 in “snow” is not one of the many fantasy stories of the late Jaworowski?
      Further you may not like the ice core data, but that are not proxies. When I see the same CO2 levels for the same time frame for ice cores with extreme differences in accumulation, temperature and impurities within +/- 5 ppmv, then the ice cores have proven themselves without any external help (be it that ocean sediments show the same variations, be it with an even worse resolution, but up to 2 million years).
      It is known (and measured) from cracks in ice cores that much higher CO2 levels ánd drilling fluid can be found. Neftel measured a lot of small samples around cracks and found a range of 260-510 ppmv for the same part of the ice. That can’t be all from the atmosphere of that time. Neftel used the values which were in line with previous and following parts, I should have rejected them all as unreliable, but I am not Neftel…
      But I don’t see that the average CO2 level from the same part of the ice core (thus the same average years) can differ that much without contamination…
      gravitational compression which forces CO2 up to the surface
      As far as I know, CO2 increase is near the bottom of stagnant air and so are the heavier isotopes, not reverse. That can be calculated from the enrichment of 15N/14N near the bottom and is about 0.45 ppmv for CO2. The same for bacterial “life” at -40°C that uses CO2 as carbon source for DNA repair: less than 0.5 ppmv at the highest dust inclusions. All peanuts compared to the 100 ppmv change over glacials and interglacials and the current 110 ppmv increase never seen before in the past 800,000 years, whatever you may think about ice core accuracy and resolution…
      At last, the Revelle factor is real and measured in the ocean surface, which makes that the ocean surface – atmosphere exchanges are very fast, but limited to 10% of the change in the atmosphere.
      The deep oceans are still far from saturated, but the exchange rate is much more limited (currently for a throughput of ~40 GtC/year only ~3 GtC more sink than source). Thus despite that the deep oceans can absorb many times the human emissions, the sink rates are too slow to get that done in a few years time…

  70. I should make it clear that those isotopic measurements by Keeling do not distinquish between anthropogenic and natural CO2, it includes biogenic CO2 as well as anthropogenic CO2. Also I am not suggesting that those measurements implying a short residence time refute the argument that anthropogenic CO2 has a long lifetime. To refute that, the Revelle Factor must first be refuted.

    • The CO2 ratio “evidence” is entirely a narrative, i.e., a plausible-sounding storyline which lacks actual proof. The ancients thought it was plausible that the Gods controlled the weather. Plausibility provides very haphazard results when applied to questions of science.

      • Bart, you know a lot more about the theory of high frequency feedbacks than me, but here I am the one who knows and you are far away from your knowledge…
        The δ13C story is as solid as adding a base to an acid: basic chemistry. As good as you can’t increase a pH by adding an acid, you can’t decrease the δ13C level in the atmosphere by adding CO2 from the oceans. Neither does vegetation, as that is a proven sink for CO2.
        That means that there is zero evidence that there is any increase of CO2 from oceans or vegetation, the two largest and fastest sources of CO2. All other known natural sources being too small or too slow (and also too high in δ13C level).
        That simply proves that your formula is mathematically (near) perfect, but physically impossible: you have no source for the extra natural CO2…

      • Not if:
        A) incoming extra flow from whatever source also is depleted in 13C and/or
        B) Diffusion of depleted CO2 into the atmosphere is like the residence/persistence time controversy – the d13C ratio is a result of diffusion of depleted 13C into the atmosphere, but this does not affect the overall persistence of a given parcel of undifferentiated CO2 in the atmosphere. Much in the same way that a small source of pollution can nevertheless poison an entire large body of water.
        What is proven is that the d13C dynamics are not like you project them to be. There is no way around the fact that you have to integrate temperature to get the phase to line up. And, once you integrate the temperature, it’s game over, because the trend in temperature takes care of the trend in the rate of change of CO2, and there is little to no room for human inputs to add to it.

      • A) An unknown other low-13C source must be fully synchronized with the human contribution which would be remarkable.
        B) Agreed, but there are only two possibilities: the high δ13C addition dilutes the δ13C “fingerprint” from the human contribution and/or it is additional to the human contribution. In the first case, the high-δ13C is only throughput and doesn’t affect the increase in the atmosphere, in the second case, the increase in the atmosphere would be 4 times higher than measured. A combination of both is -theoretically- possible if the sinks also increased a 4-fold for the combination, but that includes a 4-fold increase in throughput (as the sink rate for human and natural CO2 is the same), for which is no indication at all…
        C) See further down: there is no integration of temperature to explain the trend, variability and trend have not the slightest connection with each other.

      • The things you think are unlikely are actually very likely with feedback dynamics. Feedback systems do not behave intuitively. You have to have experience with them, and know what to look for.

    • Richard, as the δ13C variability in ice cores is not more than a few tenths of a per mil over the past 800,000 years, including glacial and interglacial periods, repeated in leaves and coralline sponges over different periods, and the drop since 1850 is ~1.6 per mil and counting and you need to burn down (without any regrowth) 1/3rd of all land vegetation to have the same effect, then you need some better arguments to show that humans are NOT the cause of the δ13C decline…
      Further, human CO2 has the same residence time as all CO2: about 5 years, but that doesn’t influence the e-fold decay time needed to reduce any excess CO2 (whatever its origin) above equilibrium.
      And please, the Revelle factor is real, measured and can be theoretically derived:
      http://www.eng.warwick.ac.uk/staff/gpk/Teaching-undergrad/es427/Exam%200405%20Revision/Ocean-chemistry.pdf

    • Richard,
      Some addition:
      The ocean surface at Bermuda (and similar at other places) shows an increase in DIC of 32 μmol in the period 1983-2012, that is an increase of 1.5% above the 2100 μmol total carbon in the ocean surface at Bermuda:
      http://tos.org/oceanography/archive/27-1_bates.pdf
      In the same period, CO2 in the atmosphere increased with 42 ppmv from a height of 343 ppmv or an increase of 12%. That gives a Revelle/buffer factor of around 8.
      Henry’s law is only for free CO2 (gas) in solution, not for (bi)carbonates: for the same temperature there is as much free CO2 in fresh water as in seawater for the same pressure of CO2 in the atmosphere. The difference is in the rest of the carbon species:
      In fresh water, it is 99% free CO2, 1% bicarbonate and virtually no carbonate
      In seawater, it is 1% free CO2, 90% bicarbonate and 9% carbonate.
      Because of the shift of equilibrium, an increase of CO2 in the atmosphere will dissolve (in this case) 8 times more CO2 in seawater than in fresh water, that is what the Revelle/buffer factor says…

  71. Willis Eschenbach January 31, 2015 at 12:02 am
    “It doesn’t work if the temperature goes stable. It doesn’t work if the temperature goes down.”
    It most certainly does. It works perfectly well since at least 1958, and can be expected to work for some to-be-observed interval into the future.
    Since the “pause”, temperatures have been stable, or even decreasing slightly. The
    dCO2/dt = k*(T – Teq)
    model tracks that interval with high fidelity. The emissions data, on the other hand, DO NOT. Emissions are accelerating. Concentration is NOT.
    rgbatduke January 31, 2015 at 7:52 am
    ‘This leads to the question (which Bart studiously avoided answering above) “What do you think the atmospheric CO_2 concentration would be if we added zero GT to the atmosphere instead of the roughly 10 corresponding to the observed increase?”’
    I’m not “studiously avoiding” anything, Doc. I’m trying to answer what seem to be the questions which will turn on the light bulb for you guys.
    What do I think the atmospheric CO_2 concentration would be if we added zero GT to the atmosphere instead of the roughly 10 corresponding to the observed increase? Almost precisely what it currently is. I think we are responsible for about the proportion of total input CO2 we produce – probably around 3% of the total.
    That’s how a feedback system works. You may know a lot of physics Doc, but it seems you too have little experience with feedback systems.
    “But if it were weeks or months it wouldn’t be anywhere near as smooth a curve as it is.”
    Of course it would. The accumulation of CO2 provides attenuation of high frequencies at -20 dB/decade. Additional smoothing takes place because of sink activity, resulting in -40 dB/decade attenuation at high frequencies. With that kind of response, it’s always going to be pretty darn smooth.
    “I also couldn’t make any sense at all out of his assertion that CO_2 doesn’t have a local equilibrium concentration or his avoidance (again) of the observation that his equation for the derivative of CO_2 concentration over time is unstable even at a constant temperature, as long as that temperature is above “T_eq”. Whatever T_eq means.”
    Stop accusing me of “avoiding” anything, Doc. Honestly, it seems like you are providing rapid fire questions, the answers to which you seem to ignore, and then are looking to accuse me of not responding to every bullet as a means of discrediting the argument.
    I gave you the equations for local equilibrium, many times:
    dCO2/dt = (CO2eq – CO2)/tau + H
    dCO2eq/dt = k*(T – Teq)
    The local equilibrium concentration is CO2eq, which is continually being shifted through temperature modulated influx of CO2 from a source as yet to be conclusively identified.
    “From my own numerical experiments with the CO_2 concentration, it is really pretty easy to create a model that fits it post 1957 based on all sorts of assumptions…”
    No, it isn’t. Not in the rate of change domain. Only the temperature model matches across the entire frequency spread.
    “…there is a very strong correspondence between anthropogenic CO_2 additions and the increase in atmospheric concentration.”
    There is a very weak correspondence. There is no match in the rate domain at all.
    Willis and RGB seem to be doing the equivalent of plugging their ears and shouting “la, la, la”. They fire questions at me, I answer them, they ignore the answers and focus on something else. It seems the point is to avoid looking at the evidence, and blithely continue to assert their own opinions based on gut feelings.

    • No Bart, you are the one that plugs your ears when the simple mass balance argument obliterates your supposed theory
      Remember, (atmospheric CO2) – (human emissions) is negative …….in spite of rising temps

      • I would surely be honored. Salby is a genius. His books are masterpieces. It is a sad commentary on our times that such an amazingly capable fellow is denigrated within his own community by mountebanks who aren’t qualified to shine his shoes, just because he is dedicated and honest enough to look at the data, and recognize where they lead.

      • Salby’s theory is garbage. Per his “theory” when global temperatures drop 2.6 degrees C, CO2 concentrations will hit zero. So, in the past glaciation, all plant life died, and neither you nor I am alive right now.

      • PS….Bart, is it true that when you work can’t pass peer review, you end up writing “books?”

  72. A note for specialists: The model
    dCO2/dt = (CO2eq – CO2)/tau + H
    dCO2eq/dt = k*(T – Teq)
    is an analogy of how the system works. Due to diffusion dynamics, there probably is not a single, dominant time constant, but a “fat tail” response. To cover that case, “tau” can be considered an operator theoretic quantity. What matters for the purposes of the discussion is the zero-frequency, “dc gain” of tau.

    • Except that CO2 according to you is tracking CO2eq near instantaneously and CO2eq is now sky high, which violates Henry’s law for the solubility of CO2 in seawater (and near all other observations)…
      Henry’s law says that CO2eq = k*(T-Teq) where k is ~8 ppmv/K no matter if that is static or dynamic..

      • “Except that CO2 according to you is tracking CO2eq near instantaneously…”
        Order of a few years, perhaps. Where is it written that this is prohibited?
        “…which violates Henry’s law…”
        Doesn’t. Not if the pCO2 of the oceans is steadily increasing from below.
        “Henry’s law says that CO2eq = k*(T-Teq)…”
        Doesn’t match the data. If you can’t match the phase response, your model is wrong.
        And, no. Henry’s Law says that CO2eq = kh(T)*pCO2(ocean)
        If pCO2(ocean) is steadily increasing, so is CO2eq. Moreover, as kh is a function of T, its rise is modulated by T.

      • Further note that Henry’s Law is for an equilibrium condition, not a dynamic one. That is why you have to integrate the temperature dependence, and that is why there is a 90 deg phase lag.

      • Bart,
        According to Wiki (sometimes reliable), Henry’s law is:
        pCO2(ocean) = kh(T) * [CO2](ocean)
        The partial pressure of CO2 from the oceans depends of the CO2 concentration in the oceans and Henry’s “constant” which changes with temperature.
        If we may start with steady state conditions, the concentration in the ocean upwelling is constant. The dependence of CO2eq on T, according to Wiki, in the following link (behind a pay wall):
        http://www.sciencedirect.com/science/article/pii/S0967064502000036
        gives a pCO2eq doubling per 16°C or about 5% per °C for CO2 in seawater.
        For the upwelling places, the maximum pCO2 measured is 750 μatm. With a pCO2 difference of ~350 μatm between ocean and atmosphere (at ~400 μatm), that gives an influx of about 40 GtC/year.
        If the ocean temperature increases with 1°C, the pCO2 at the upwelling places increases with 5% to 787.5 μatm. That makes that the CO2 pressure difference increases from ~350 μatm to ~388 μatm and as the influx is directly proportional to the pressure difference, the CO2 influx increases from 40 GtC/year to 44 GtC/year.
        The same situation for a 10% increase in CO2 concentration: that gives a 10% increase in pCO2(ocean) and therefore an increase of the influx to 48 GtC/year.
        The combination of both has a small quadratic term in it, then the overall increase in influx is 53.3 GtC/year.
        Here your story stops, completely ignoring the feedback from the increase of CO2 pressure in the atmosphere.
        According to you:
        dCO2eq/dt = k*(T – Teq), the CO2eq of the oceans with the atmosphere changes over time. According to Henry’s law, CO2eq is temperature dependent but is fixed: there is a unique pCO2(ocean) for a unique change in T and CO2 concentration, which initially gives a maximum CO2 influx, but as the pCO2 in the atmosphere increases, the CO2 input reduces and ultimately is zero for a pCO2 in the atmosphere of 750 μatm (original), 787.5 μatm (T-increase), 825 μatm (concentration increase) or 866 μatm for the combination.
        With other words, there is no constant CO2 influx for a sustained step change in temperature and/or concentration. And there is not the slightest indication for a continuous increase of concentration and/or total upwelling.
        In reality, the changes in influx are smaller, as the above changes are calculated for the maximum measured pCO2. And there are also changes in outflux: as the pCO2 in the atmosphere increases, the pCO2 difference between atmosphere and oceans increases, pushing more CO2 into the ocean sink places. The net effect is a new equilibrium at about halve the change of input flux and/or ~8 ppmv/°C (4-17 according to the literature, 8 ppmv/°C over glacial-interglacial intervals).
        Again, the variability of the CO2 trend has nothing to do with the oceans, that is proven beyond doubt: it is caused by the reaction of vegetation to fast temperature changes. But vegetation is not the cause of the trend, based on the oxygen balance. Neither are the oceans as they have a too high δ13C level… Variability and trend are completely independent processes and you can’t deduce anything about the cause of the trend by looking at the variability around the trend…

      • The change in pCO2 of the oceans would necessarily be precisely what is needed to balance with the pCO2 of the atmosphere whether the atmospheric increase were human induced or not, so your assertions fall flat here. The increase in pCO2 of the oceans can easily be enough to drive the atmospheric change.
        The data tell us that temperature must be integrated to achieve a match with the CO2 data. You must match the phase.
        When you do that, it is apparent that human inputs are superfluous.
        Moreover, it explains the current stall in the rate of change of atmospheric CO2 coincident with the stall in temperatures.
        There is no way around it, no doubt about it. Human inputs are not the driver.

      • Come on Bart, all you have is a simple curve fit for the trend which has no bearing in any known physical process.
        Your formula:
        dCO2eq/dt = k*(T – Teq)
        does imply a continuous increase of CO2eq for a one-step change in temperature, while Henry’s law shows a fixed step change of CO2eq for a fixed step change in temperature (as good as for a step change in concentration) and completely ignores the feedback from the increased pCO2 in the atmosphere on input and output fluxes of the ocean…

      • No, it shows a change in the rate at which increased atmospheric concentration is occurring. It is not the temperature which keeps the CO2 coming in, it is the k factor, which represents an established flow which is temperature dependent.
        You have
        CO2 = kh(T)*pCO2(oceans)
        Now, assume pCO2(oceans) is increasing. Assume the rate of change of kh(T) is relatively slow, so we have
        dCO2/dt := kh(T)*dpCO2(oceans)/dt
        A linear approximation of kh(T) is kh(T) = k0 + k1*T, so
        dCO2/dt := (k0 + k1*T)*dpCO2(oceans)/dt
        Assuming dpCO2(oceans)/dt is fairly steady, set k = k1*dpCO2(oceans)/dt and Teq = -k0/k1. Then, the equation becomes
        dCO2/dt := k*(T – Teq)
        It’s just a mathematical expression describing a temperature dependent influx. Nothing far-fetched about it.

      • Bart,
        Now you are mixing two separate points:
        – a steadily increasing concentration of CO2 in the ocean upwelling
        – a fixed increase in temperature.
        Your end formula is mathematically right, but the bulk of the increase is from the steadily increasing CO2 concentration in upwelling. Not from the temperature increase…
        Your last formula shows the end result where the increase rate k is including the factor caused by the increase in upwelling for a fixed step change in temperature. That needs a different k for a different rate of change of the upwelling. Besides a small quadratic term, near all of the increase in dCO2/dt is from the steadily increasing upwelling, not from the step increase in temperature. If there was no steadily increase in upwelling, only a step change, there would be a fixed change of CO2 in the atmosphere, not the steadily increasing increase noticed in the past 55 years.
        Thus your formula wrongly attributes the increase in CO2 to temperature, while the real increase is from the upwelling…
        Besides that, there is not the slightest observation of an increase in upwelling, which should be a 4-fold over the past 55 years to beat the 4-fold increase of human emissions in that period…

  73. Ferdinand, this is nothing different than what I have been arguing for years. The formula doesn’t “wrongly attribute” anything. You do. You haven’t been paying attention to my arguments. Now that you see it, maybe you can give it some serious thought.
    Yes, there is evidence of increased temperature dependent input from some source. It is the evidence that you have to integrate the temperature to get the proper phasing. It is the evidence that emissions are currently accelerating, while atmospheric concentration is not.

    • Sorry Bart,
      Your formula hardly depends on temperature, it depends (near) entirely on the continuous increasing increase in upwelling, with only a small quadratic component where the temperature increase has some influence.
      Almost all of the increase is hidden in the factor k which mainly depends on the upwelling, and a little on temperature…

      • So what? As I say, this is what I have been arguing all along. You can see it, for example, here from over two years ago.
        The temperature is a critical component. It is what produces the slope in dCO2/dt. Human emissions also have a slope in the rate of change. Since it is already accounted for, there is little to no room to add in human emissions.

      • I might should say “trend” rather than “slope”. Not much difference – a sustained slope is a trend.

      • Bart, no problem with slope or trend, I use them both too.
        According to your work of two years ago (where I reacted too):
        CO2(boundary) = Kh*CO2_Oceans(boundary)
        The derivative of this is
        dCO2(boundary)/dt = dKh/dt*CO2_Oceans(boundary) + Kh*dCO2_Oceans(boundary)/dt
        Kh is a function of temperature, and thus can be expanded to first order as
        Kh = Kh_eq + Kh_partial*(T – Teq)
        where Kh_partial is the partial derivative of Kh to temperature. The oceans have been a net source of CO2 to the atmosphere. Assuming these are dominant, then
        dCO2(boundary)/dt := (Kh_partial*dCO2_Oceans(boundary)/dt) * (T – Teq)

        Besides the (measured) fact that the oceans are a net sink for CO2, not a source (as HS and you wrongly interpreted the partial sink/source exchange with the atmosphere), where is said that dCO2_Oceans/dt increases over time? And where – again – is the influence of CO2 in the atmosphere (and dCO2(atm)/dt), as that increases over time? The CO2 influx doesn’t depend of CO2_oceans alone, it depends of the difference between CO2_oceans and CO2_atmosphere…
        Again, a step change in temperature and/or pCO2(oceans) at the upwelling side will give you a finite increase of CO2 in the atmosphere. Only a continuous increasing increase in temperature and/or concentration will give you a continuous increase of CO2 in the atmosphere. There is not the slightest indication that the ocean influx or CO2 concentration changed at the upwelling places.

      • From above:

        Yes, there is evidence of increased temperature dependent input from some source. It is the evidence that you have to integrate the temperature to get the proper phasing. It is the evidence that emissions are currently accelerating, while atmospheric concentration is not.

        You’re not making any new argument, Ferdinand. You’re just digging in your heels, and insisting we should not look at the evidence.
        You must fit all components, with the proper phasing, in order to have a viable model. As of right now, you do not have a model which fits the data. I do.

  74. Bart,
    It is you who shifted the argument from “temperature is the sole factor” in the slope of the rate of change to “the oceans show an increasing rate of upwelling”, hidden in the definition of k. In the latter case, it is the increase rate of the upwelling which is responsible for the bulk of the increase rate of CO2, not temperature, which is only a small part of the increasing rate of change and hardly plays a role: the slope would be near the same if there was no temperature increase at all.
    And your “evidence” is no evidence at all, as the phasing is from the “noise” caused by temperature, but the slope has nothing to do with the variability around it (which is from vegetation) and near nothing with temperature…

    • As I documented with the link above, I have been making this argument for at least two years, and actually much longer. It is not my fault if you have been ignoring what I have been writing, and making up a straw man to argue against.
      And, no. The phasing is there for a very definite reason – temperatures are modulating the rate of change of CO2. That means CO2 evolves as the integral of temperature anomaly. And, that means the curvature in the plot of CO2 is accounted for, and human emissions cannot be the driving force.
      When you have been doing system identification as long as I have, you learn that you cannot just dismiss less significant, but still significant and, what is more, repeatable, details as “noise”. It is those details which tell you most about the system you are trying to identify. If you cannot explain them, then your model is not going to work.

      • PS: I am not specifically arguing that “the oceans show an increasing rate of upwelling”. I am arguing that, that is a likely candidate for what we are seeing.
        Whether it is from ocean upwelling or not, there is some temperature dependent source of influx which is driving the observations. Human emissions are not temperature dependent. Ergo, they are not what is driving it.

      • Bart, the temperatures are modulating the rate of change of CO2, but only for the processes which are directly affected by temperature: mainly tropical vegetation. That is proven beyond doubt.
        There is no reason to integrate T for the slope part, as you don’t know the reason for the slope, which is anyway not from vegetation.
        By integrating the whole temperature trend with an huge factor (for which is no physical explanation), you include the slope of the derivative as part of a temperature driven process, while you have no proof at all that it is temperature driven. That is circular reasoning: you exclude other probable factors based on an arbitrary choice.
        Further, your argument of two years ago was:
        dCO2(boundary)/dt := (Kh_partial*dCO2_Oceans(boundary)/dt) * (T – Teq)
        Nowhere is said that dCO2_Oceans(boundary)/dt is increasing over time. If dCO2_Oceans(boundary)/dt is constant, then there is no slope in the overall derivative with a constant (T-Teq), only if you use an arbitrary factor and offset for (T-Teq), which has no bearing in any known physical process.
        But there is a fourfold increase in the derivative since 1960, which is only possible either by human emissions, which show a fourfold increase in dCO2(emissions)/dt or alternatively by a fourfold increasing dCO2_Oceans/dt over time, where T-Teq is a very small extra factor in the slope.
        And as usual, you completely ignore the feedback on sources and sinks from the increased CO2 pressure in the atmosphere…

      • You must integrate the entire temperature record. For a low frequency cutoff, there would have to be a high pass filtering operation going on, which would leave a distinctive phase distortion, which is not evident.
        When you fit the variation, the trend fits, too. Occam’s razor – it all fits, and is the simplest explanation for what it going on.

  75. OK. let us give a practical example based on real life figures:
    The current CO2 flux at the upwelling sites is ~40 GtC/year, rough estimate based on 13C and 14C “dilution” rates.
    An increase of 1°C gives a 5% increase in ocean pCO2.
    An increase of 10% in CO2 concentration in the ocean upwelling gives a 10% increase in ocean pCO2.
    A combination of both gives a 15.5% increase in ocean pCO2 (the 0.5% extra is responsible for your “slope” in the derivative!)
    The increase in temperature 1960-2012 was ~0.7°C. That gives an increase of 3.5% in ocean pCO2.
    The increase in CO2 of the atmosphere was slightly quadratic with about 80 ppmv
    The rate of change of CO2 increased over the same period from 0.5 ppmv/year to 2 ppmv/year
    (BTW WoodforTrees still calculates the rate of change per month, whatever the sample rate chosen)
    For the given increase of CO2 over time, you need an initial influx difference between sources and sinks of ~1 GtC/year in 1960, increasing to ~4 GtC/year in 2012.
    Assuming that the sinks are constant, the influx need to (exponentially!) increase from ~41 GtC/year to ~44 GtC/year between 1960 and 2012, against an increase of pCO2 in the atmosphere of ~80 ppmv.
    The maximum pCO2(ocean) measured was 750 μatm (reference year 1995). Assuming that the pCO2(oceans) started at ~700 μatm in 1960, the pCO2 difference between oceans and atmosphere then was 700-315 = 385 μatm for a 41 GtC/year influx.
    To increase the influx from 41 to 44 GtC one need an increase of the pCO2 difference of 385*44/41=413 μatm.
    As the pCO2(atmosphere) in 2012 was increased with 80 ppmv to 395 ppmv, pCO2(oceans) need to increase to ~800 μatm to reach the 44 GtC/year influx. Or an increase of 14.3%. The temperature part of that increase is 3.5 μatm (800*3.5% – 700*3.5%) or about 3.5% of the total increase in μatm or ppmv or rate of change…
    Thus if the cause of the increase in the atmosphere was ocean upwelling, practically all increase is from the upwelling itself and the temperature increase has a very small, practically negligible influence.
    If the increase in sink rate by the increased pCO2(atmosphere) is taken into account, then the increase in CO2 concentration in the upwelling must double to have the same effect, but that doesn’t change the small temperature influence.

    • You’re just handwaving. The increase in pCO2 of the oceans would be exactly what was needed to support the increase in CO2 in the atmosphere, regardless of whether human inputs were forcing an increase in the atmosphere forcing a pCO2 change in the oceans, or the oceans were upwelling to force the pCO2 change in the atmosphere.
      You can’t get to attribution along this path.
      What tells us it isn’t human forcing is the fact that the human forcing is already linear in the rate of change, and the rate of rise in concentration would become quadratic with the modulation of sink activity by temperatures. But, the rate of rise in concentration has been at best linear, and is currently in a stall with the stall in temperatures. Moreover, the variations would not match, as they would be becoming progressively greater in amplitude, and they are steady, and match the variations in temperature.
      The form of the driving input must match the form of the output. Human inputs simply would not produce the output atmospheric concentration we are seeing if they were in the driver’s seat.

      • Bart,
        The CO2 variations do match the temperature variations and are caused by vegetation. Human emissions have no influence on the amplitude of the variability, as there is hardly any influence of more CO2 on temperature, but human emissions have an influence on the CO2 uptake by vegetation: about 10% of all human emissions is extra absorbed in more vegetation growth than decay.
        The CO2 trend is from a different process, no matter if that is from the oceans or humans. Human emissions are not influenced by temperature, ocean emissions hardly. In both cases, temperature is not the driving force of the increase. And ocean emissions can’t be the cause because of a too high δ13C level.
        Thus the increase in the atmosphere is the sum of two independent processes: temperature for the variability(which gives a very small contribution to the CO2 increase when integrated) and human emissions for the increase in the atmosphere minus the sink rate which is in ratio to the increase in the atmosphere compared to the temperature influenced equilibrium.
        And as plotted many times for you, the increase in the atmosphere caused by human emissions is largely within natural variability, which for yearly averages is between 10% and 90% of the emissions and average 50-55%, including longer periods of standstill and accelerating. The calculated increase in the atmosphere is the red line in:
        http://www.ferdinand-engelbeen.be/klimaat/klim_img/dco2_em4.jpg
        Thus human emissions trend + temperature caused variability match the form and the trend of the increase in the atmosphere…

      • No, Ferdinand. The phase does not match with temperature and CO2. You must match the phase.
        And, your plot is a worthless least squares fit. The fact remains that temperatures are at a steady rate, and emissions are accelerating, and even your fit cannot hide that.

  76. It is my belief that CO2 levels are determined by the equilibrium level enforced by the ocean atmospheric interface. Salby, I have read, thinks that soil respiration, which is poorly quantified, could be the controlling interface. Perhaps it is both. Or, perhaps there are other processes of which we are currently unaware.
    But, the bottom line is that the relatively small amounts of human inputs are rapidly sequestered, and overall atmospheric content is determined by the natural interfaces. These interfaces enforce an attractor level, which I loosely call an equilibrium level though it is slowly time varying.
    Human inputs cannot be the driver. That putative input does not match the form of the output.

    • Perhaps I should have said “that putative driving input”. It is obviously an input. The question is whether it is the driver. The data say it isn’t.

    • Bart,
      The short term (seasons to 2-3 years) drivers of CO2 in the atmosphere is the influence of temperature on vegetation. That are very fast to fast responses of growing and decaying leaves, including the soil bacteria of Dr. Salby. Both show about 4-5 ppmv/K change.
      The pre-industrial equilibrium was largely caused by the influence of temperature on ice sheets, ocean surface and especially deep oceans (upwelling and downwelling). That was about 8 ppmv/K over extremely long periods.
      The current increase is not from vegetation: that would need to burn down 1/3rd of all land vegetation without regrowth to have the same effect. It is not from the oceans, as the δ13C level is too high (and they are a measured, limited net sink for CO2).
      The point is that the human input is NOT easily sequestered, as most natural processes have a rather fixed seasonal or longer term input and/or output, which can be seen in the quite small year by year variability of +/- 2 GtC/year while the total natural seasonal in/out fluxes are estimated of around 150 GtC. That is a natural variability of only +/- 1.3% of the main fluxes, while the human input is ~9 GtC/year or 6% of the main fluxes…

      • Mere assertions, all.
        The data are clear. You’ve got to match the phase. You’ve got to account for the stall in CO2 rate of change even as emissions accelerate.

      • Bart,
        CO2 phase doesn’t match T phase: CO2 changes follow T changes with 90 deg. It should follow, as an increase in temperature does immediately increase the influx rate, but reaching a higher CO2 level needs time. So does the derivative of CO2 follow the derivative of T with 90 deg. Your match of phases proves that dCO2/dt is not caused by T…
        Besides that, it is impossible that a small fixed change in T will give a continuous increase of CO2 in the atmosphere from the oceans without feedback from the increased pressure in the atmosphere…

      • And, when you have a 90 deg phase lag across all observable frequencies with no discernible phase lag for a given time interval, that means that you effectively have an integration across that timeline. There is no way around it.
        The phase lag of 90 deg means that you must integrate that trend in temperature to get CO2. And, when you do that, there is little to no room left for anthropogenic attribution.
        “Besides that, it is impossible that a small fixed change in T will give a continuous increase of CO2…”
        You’re backtracking. We’ve already established one potential mechanism by which it can, when pCO2(oceans) is increasing due to upwelling of CO2 rich waters.

      • Yes Bart,
        You have to integrate the temperature to obtain the CO2 variability but only for that part that shows the variability, as that is a fast process. Not for the trend, which is a much slower process and has nothing to do with the fast responses to temperature. You still mix both responses as if it was caused by one process…
        The net result is the sum of two “natural” integrals: the fast one which gives the fast responses (vegetation) and the slow changing (oceans) which may give a response on (very) long time frames, plus the integrated human contribution minus the response of the sinks to the CO2 pressure increase in the atmosphere, which you always seem to forget…
        The integral from the effect of temperature variations on vegetation is essentially zero to slightly negative over the past 55 years. The integral of the influence of temperature on the oceans since 1960 minus the response from the increased CO2 pressure caused by temperature is about 6 ppmv… The rest is from human emissions…
        As this is near the bottom of the main page, see you next round…

      • No, Ferdinand. You are claiming a high pass filter operation on the temperature which would leave visible phase distortion. There is no visible phase distortion.

      • Remind me:
        How do we know that the human input is not easily sequestered ? As a little polyp on a south pacific reef, anytime there is excess temperature and available carbonate, I’m a happy little chappy.
        and thats just me. my distant cousins the phytoplankton are more than happy with some extra near surface enthalpy
        references please

  77. A small addition:
    You’re backtracking. We’ve already established one potential mechanism by which it can, when pCO2(oceans) is increasing due to upwelling of CO2 rich waters.
    Sorry Bart, that is only possible for an increasing increase of pCO2(oceans) in the oceans (which doesn’t match the 13C/12C ratio decline). Very little (3.5%) to do with an increase in temperature. Thus hardly any influence of a sustained increase in temperature as your formula says, as near all increase is from the steadily increasing factor k from your upwelling…

    • Doesn’t matter how small you think it is, it is significant, and you have to match it. The data almost perfectly match the model dCO2/dt = k*(T – Teq) across the entire range of observed frequencies. There is no doubt about it.
      This model is consistent with steadily increasing pCO2(oceans) from a source not anthropogenic. Anthropogenic forcing as a driving input is not consistent with the atmospheric CO2 observations.